Token | Can |
Token ID | 6090 |
Feature activation | -3.71e-02 |
Pos logit contributions | |
hotel | +0.017 |
principal | +0.017 |
ey | +0.017 |
city | +0.017 |
classroom | +0.016 |
Neg logit contributions | |
es | -0.013 |
icked | -0.012 |
on | -0.012 |
together | -0.011 |
there | -0.011 |
Token | you |
Token ID | 345 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
es | +0.043 |
icked | +0.038 |
on | +0.037 |
together | +0.036 |
there | +0.034 |
Neg logit contributions | |
hotel | -0.094 |
principal | -0.091 |
city | -0.090 |
classroom | -0.090 |
ey | -0.090 |
Token | help |
Token ID | 1037 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
hotel | +0.018 |
ey | +0.017 |
classroom | +0.017 |
city | +0.017 |
principal | +0.017 |
Neg logit contributions | |
es | -0.023 |
icked | -0.021 |
on | -0.021 |
together | -0.021 |
there | -0.021 |
Token | us |
Token ID | 514 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
es | +0.045 |
icked | +0.043 |
on | +0.040 |
together | +0.039 |
there | +0.038 |
Neg logit contributions | |
hotel | -0.042 |
principal | -0.041 |
museum | -0.040 |
city | -0.040 |
classroom | -0.040 |
Token | fix |
Token ID | 4259 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
hotel | +0.023 |
classroom | +0.022 |
city | +0.022 |
museum | +0.022 |
principal | +0.022 |
Neg logit contributions | |
es | -0.033 |
icked | -0.032 |
on | -0.028 |
together | -0.028 |
there | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
es | +0.013 |
icked | +0.012 |
on | +0.011 |
together | +0.011 |
there | +0.011 |
Neg logit contributions | |
hotel | -0.026 |
city | -0.025 |
principal | -0.025 |
classroom | -0.025 |
ey | -0.025 |
Token | train |
Token ID | 4512 |
Feature activation | +5.12e-03 |
Pos logit contributions | |
ey | +0.297 |
igsaw | +0.268 |
former | +0.266 |
Ta | +0.260 |
priest | +0.260 |
Neg logit contributions | |
on | -0.405 |
there | -0.371 |
in | -0.348 |
together | -0.341 |
at | -0.327 |
Token | set |
Token ID | 900 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
hotel | +0.012 |
city | +0.012 |
classroom | +0.011 |
principal | +0.011 |
museum | +0.011 |
Neg logit contributions | |
es | -0.007 |
icked | -0.006 |
on | -0.006 |
together | -0.006 |
there | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | +3.14e-03 |
Pos logit contributions | |
es | +0.022 |
icked | +0.021 |
together | +0.017 |
there | +0.016 |
on | +0.016 |
Neg logit contributions | |
hotel | -0.037 |
classroom | -0.035 |
principal | -0.035 |
airport | -0.035 |
city | -0.035 |
Token | please |
Token ID | 3387 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
hotel | +0.008 |
ey | +0.008 |
city | +0.008 |
classroom | +0.008 |
principal | +0.008 |
Neg logit contributions | |
es | -0.004 |
on | -0.003 |
together | -0.003 |
there | -0.003 |
icked | -0.003 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
es | +0.055 |
icked | +0.050 |
led | +0.042 |
together | +0.041 |
on | +0.040 |
Neg logit contributions | |
museum | -0.060 |
hotel | -0.060 |
classroom | -0.060 |
principal | -0.059 |
city | -0.058 |
Token | I |
Token ID | 314 |
Feature activation | +1.97e-03 |
Pos logit contributions | |
es | +0.006 |
on | +0.006 |
icked | +0.005 |
together | +0.005 |
there | +0.005 |
Neg logit contributions | |
hotel | -0.010 |
ey | -0.010 |
classroom | -0.010 |
city | -0.010 |
principal | -0.010 |
Token | hope |
Token ID | 2911 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
hotel | +0.004 |
ey | +0.004 |
principal | +0.004 |
city | +0.004 |
classroom | +0.004 |
Neg logit contributions | |
es | -0.003 |
on | -0.003 |
together | -0.003 |
there | -0.003 |
icked | -0.003 |
Token | you |
Token ID | 345 |
Feature activation | +1.74e-02 |
Pos logit contributions | |
es | +0.029 |
icked | +0.026 |
on | +0.024 |
together | +0.024 |
there | +0.023 |
Neg logit contributions | |
hotel | -0.049 |
principal | -0.047 |
classroom | -0.047 |
city | -0.047 |
museum | -0.047 |
Token | can |
Token ID | 460 |
Feature activation | +5.36e-03 |
Pos logit contributions | |
hotel | +0.035 |
city | +0.033 |
classroom | +0.032 |
playground | +0.032 |
museum | +0.032 |
Neg logit contributions | |
es | -0.028 |
icked | -0.026 |
on | -0.026 |
together | -0.025 |
there | -0.025 |
Token | fix |
Token ID | 4259 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
hotel | +0.011 |
ey | +0.010 |
city | +0.010 |
principal | +0.010 |
classroom | +0.010 |
Neg logit contributions | |
es | -0.009 |
on | -0.009 |
together | -0.008 |
there | -0.008 |
icked | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
es | +0.020 |
icked | +0.018 |
on | +0.018 |
together | +0.017 |
there | +0.017 |
Neg logit contributions | |
hotel | -0.049 |
ey | -0.048 |
city | -0.047 |
classroom | -0.047 |
principal | -0.047 |
Token | file |
Token ID | 2393 |
Feature activation | -2.17e-02 |
Pos logit contributions | |
hotel | +0.313 |
orchestra | +0.295 |
priest | +0.292 |
actor | +0.280 |
former | +0.280 |
Neg logit contributions | |
on | -0.364 |
there | -0.340 |
in | -0.322 |
at | -0.317 |
es | -0.301 |
Token | ." |
Token ID | 526 |
Feature activation | -6.69e-03 |
Pos logit contributions | |
icked | +0.020 |
es | +0.020 |
on | +0.015 |
there | +0.015 |
together | +0.014 |
Neg logit contributions | |
hotel | -0.058 |
city | -0.055 |
museum | -0.055 |
classroom | -0.055 |
principal | -0.055 |
Token |
|
Token ID | 198 |
Feature activation | +1.88e-03 |
Pos logit contributions | |
es | +0.010 |
icked | +0.009 |
on | +0.009 |
together | +0.009 |
there | +0.008 |
Neg logit contributions | |
ey | -0.014 |
classroom | -0.014 |
hotel | -0.014 |
principal | -0.014 |
city | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -2.06e-03 |
Pos logit contributions | |
hotel | +0.004 |
ey | +0.004 |
classroom | +0.004 |
principal | +0.004 |
city | +0.004 |
Neg logit contributions | |
es | -0.003 |
icked | -0.002 |
on | -0.002 |
together | -0.002 |
there | -0.002 |
Token | Mom |
Token ID | 29252 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
es | +0.003 |
icked | +0.003 |
on | +0.003 |
together | +0.003 |
there | +0.003 |
Neg logit contributions | |
hotel | -0.004 |
ey | -0.004 |
classroom | -0.004 |
principal | -0.004 |
city | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | +4.30e-02 |
Pos logit contributions | |
es | +0.027 |
icked | +0.024 |
on | +0.024 |
together | +0.024 |
there | +0.023 |
Neg logit contributions | |
hotel | -0.040 |
ey | -0.038 |
classroom | -0.038 |
city | -0.038 |
principal | -0.038 |
Token | bit |
Token ID | 1643 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
ey | +0.092 |
opl | +0.086 |
orchestra | +0.082 |
Ta | +0.078 |
principal | +0.078 |
Neg logit contributions | |
on | -0.075 |
there | -0.073 |
together | -0.072 |
, | -0.067 |
in | -0.066 |
Token | of |
Token ID | 286 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
es | +0.024 |
on | +0.022 |
icked | +0.022 |
together | +0.022 |
there | +0.021 |
Neg logit contributions | |
hotel | -0.029 |
ey | -0.028 |
city | -0.028 |
classroom | -0.028 |
principal | -0.028 |
Token | trouble |
Token ID | 5876 |
Feature activation | -4.41e-02 |
Pos logit contributions | |
hotel | +0.045 |
principal | +0.044 |
ey | +0.044 |
priest | +0.044 |
orchestra | +0.043 |
Neg logit contributions | |
on | -0.051 |
together | -0.050 |
there | -0.049 |
instead | -0.045 |
it | -0.045 |
Token | fixing |
Token ID | 18682 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
es | +0.052 |
icked | +0.048 |
on | +0.043 |
together | +0.043 |
there | +0.041 |
Neg logit contributions | |
hotel | -0.110 |
city | -0.106 |
classroom | -0.105 |
playground | -0.105 |
principal | -0.104 |
Token | the |
Token ID | 262 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
es | +0.022 |
on | +0.020 |
icked | +0.020 |
together | +0.019 |
there | +0.019 |
Neg logit contributions | |
hotel | -0.037 |
ey | -0.036 |
city | -0.035 |
classroom | -0.035 |
principal | -0.035 |
Token | bag |
Token ID | 6131 |
Feature activation | -4.28e-02 |
Pos logit contributions | |
orchestra | +0.293 |
ey | +0.288 |
actor | +0.282 |
priest | +0.277 |
former | +0.276 |
Neg logit contributions | |
on | -0.351 |
there | -0.340 |
in | -0.309 |
at | -0.299 |
with | -0.299 |
Token | , |
Token ID | 11 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
es | +0.043 |
icked | +0.039 |
on | +0.034 |
together | +0.033 |
there | +0.032 |
Neg logit contributions | |
hotel | -0.113 |
city | -0.110 |
classroom | -0.109 |
playground | -0.109 |
principal | -0.108 |
Token | but |
Token ID | 475 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
es | +0.033 |
on | +0.028 |
together | +0.027 |
icked | +0.027 |
there | +0.027 |
Neg logit contributions | |
hotel | -0.090 |
ey | -0.088 |
city | -0.087 |
classroom | -0.087 |
principal | -0.087 |
Token | eventually |
Token ID | 4191 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
es | +0.016 |
icked | +0.013 |
on | +0.013 |
together | +0.012 |
there | +0.012 |
Neg logit contributions | |
hotel | -0.047 |
ey | -0.046 |
city | -0.046 |
classroom | -0.046 |
principal | -0.045 |
Token | he |
Token ID | 339 |
Feature activation | -7.89e-03 |
Pos logit contributions | |
es | +0.027 |
icked | +0.025 |
together | +0.022 |
iled | +0.021 |
on | +0.020 |
Neg logit contributions | |
hotel | -0.052 |
beach | -0.049 |
city | -0.049 |
museum | -0.048 |
classroom | -0.048 |
Token | worry |
Token ID | 5490 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
hotel | +0.024 |
ey | +0.023 |
city | +0.023 |
principal | +0.023 |
classroom | +0.023 |
Neg logit contributions | |
es | -0.026 |
on | -0.024 |
together | -0.024 |
there | -0.024 |
icked | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | -8.69e-03 |
Pos logit contributions | |
icked | +0.017 |
es | +0.016 |
ged | +0.014 |
pped | +0.014 |
bled | +0.013 |
Neg logit contributions | |
hotel | -0.028 |
city | -0.027 |
classroom | -0.027 |
difference | -0.027 |
museum | -0.027 |
Token | we |
Token ID | 356 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
es | +0.009 |
icked | +0.008 |
together | +0.006 |
on | +0.006 |
there | +0.005 |
Neg logit contributions | |
hotel | -0.023 |
city | -0.023 |
ey | -0.022 |
classroom | -0.022 |
playground | -0.022 |
Token | will |
Token ID | 481 |
Feature activation | -4.83e-03 |
Pos logit contributions | |
hotel | +0.041 |
ey | +0.040 |
principal | +0.039 |
classroom | +0.039 |
city | +0.039 |
Neg logit contributions | |
es | -0.030 |
on | -0.028 |
together | -0.027 |
icked | -0.027 |
there | -0.027 |
Token | fix |
Token ID | 4259 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
es | +0.009 |
on | +0.008 |
together | +0.008 |
there | +0.008 |
icked | +0.008 |
Neg logit contributions | |
hotel | -0.009 |
ey | -0.009 |
principal | -0.008 |
city | -0.008 |
classroom | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
es | +0.016 |
on | +0.013 |
icked | +0.013 |
together | +0.013 |
there | +0.012 |
Neg logit contributions | |
hotel | -0.058 |
ey | -0.057 |
city | -0.057 |
classroom | -0.056 |
principal | -0.056 |
Token | sandbox |
Token ID | 35204 |
Feature activation | -4.54e-02 |
Pos logit contributions | |
principal | +0.321 |
ey | +0.318 |
Ta | +0.317 |
orchestra | +0.316 |
hotel | +0.306 |
Neg logit contributions | |
on | -0.318 |
there | -0.306 |
in | -0.296 |
together | -0.275 |
with | -0.271 |
Token | and |
Token ID | 290 |
Feature activation | -1.03e-03 |
Pos logit contributions | |
es | +0.041 |
icked | +0.038 |
on | +0.033 |
there | +0.031 |
together | +0.030 |
Neg logit contributions | |
hotel | -0.125 |
city | -0.120 |
classroom | -0.120 |
principal | -0.119 |
playground | -0.119 |
Token | make |
Token ID | 787 |
Feature activation | -2.61e-03 |
Pos logit contributions | |
es | +0.002 |
on | +0.002 |
there | +0.002 |
together | +0.002 |
instead | +0.001 |
Neg logit contributions | |
hotel | -0.002 |
ey | -0.002 |
principal | -0.002 |
city | -0.002 |
classroom | -0.002 |
Token | it |
Token ID | 340 |
Feature activation | -3.65e-05 |
Pos logit contributions | |
es | +0.002 |
together | +0.002 |
on | +0.002 |
there | +0.002 |
icked | +0.002 |
Neg logit contributions | |
hotel | -0.007 |
ey | -0.007 |
classroom | -0.007 |
principal | -0.007 |
city | -0.007 |
Token | nice |
Token ID | 3621 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
es | +0.000 |
on | +0.000 |
icked | +0.000 |
together | +0.000 |
there | +0.000 |
Neg logit contributions | |
hotel | -0.000 |
ey | -0.000 |
city | -0.000 |
classroom | -0.000 |
principal | -0.000 |
Token | find |
Token ID | 1064 |
Feature activation | -5.74e-03 |
Pos logit contributions | |
hotel | +0.010 |
ey | +0.010 |
principal | +0.010 |
city | +0.010 |
classroom | +0.009 |
Neg logit contributions | |
es | -0.009 |
together | -0.009 |
on | -0.009 |
there | -0.009 |
instead | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | +8.26e-02 |
Pos logit contributions | |
es | +0.008 |
on | +0.008 |
together | +0.008 |
there | +0.008 |
icked | +0.007 |
Neg logit contributions | |
hotel | -0.013 |
ey | -0.013 |
principal | -0.012 |
classroom | -0.012 |
city | -0.012 |
Token | way |
Token ID | 835 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
ey | +0.136 |
hotel | +0.123 |
Ta | +0.123 |
principal | +0.120 |
orchestra | +0.118 |
Neg logit contributions | |
on | -0.179 |
there | -0.176 |
together | -0.176 |
with | -0.164 |
in | -0.164 |
Token | to |
Token ID | 284 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
es | +0.019 |
icked | +0.017 |
on | +0.014 |
together | +0.014 |
there | +0.013 |
Neg logit contributions | |
hotel | -0.047 |
city | -0.046 |
classroom | -0.046 |
playground | -0.045 |
ey | -0.045 |
Token | fix |
Token ID | 4259 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
hotel | +0.047 |
ey | +0.045 |
principal | +0.044 |
city | +0.044 |
classroom | +0.044 |
Neg logit contributions | |
es | -0.043 |
on | -0.042 |
together | -0.041 |
there | -0.041 |
instead | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
es | +0.015 |
icked | +0.013 |
on | +0.013 |
together | +0.013 |
there | +0.013 |
Neg logit contributions | |
hotel | -0.037 |
ey | -0.036 |
city | -0.036 |
classroom | -0.036 |
principal | -0.036 |
Token | situation |
Token ID | 3074 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
hotel | +0.269 |
principal | +0.268 |
babys | +0.254 |
lobby | +0.252 |
pastor | +0.250 |
Neg logit contributions | |
on | -0.386 |
there | -0.385 |
in | -0.365 |
it | -0.341 |
with | -0.340 |
Token | . |
Token ID | 13 |
Feature activation | -2.84e-03 |
Pos logit contributions | |
icked | +0.025 |
es | +0.023 |
on | +0.018 |
there | +0.017 |
together | +0.017 |
Neg logit contributions | |
hotel | -0.069 |
city | -0.067 |
classroom | -0.067 |
museum | -0.066 |
playground | -0.066 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +9.85e-03 |
Pos logit contributions | |
es | +0.004 |
on | +0.003 |
icked | +0.003 |
together | +0.003 |
there | +0.003 |
Neg logit contributions | |
hotel | -0.007 |
ey | -0.007 |
city | -0.007 |
classroom | -0.007 |
principal | -0.007 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.74e-02 |
Pos logit contributions | |
hotel | +0.017 |
city | +0.016 |
museum | +0.016 |
classroom | +0.016 |
playground | +0.015 |
Neg logit contributions | |
icked | -0.019 |
es | -0.018 |
able | -0.017 |
together | -0.017 |
instead | -0.017 |
Token | upon |
Token ID | 2402 |
Feature activation | -7.57e-03 |
Pos logit contributions | |
hotel | +0.067 |
ey | +0.064 |
city | +0.063 |
classroom | +0.063 |
principal | +0.063 |
Neg logit contributions | |
es | -0.071 |
on | -0.066 |
icked | -0.065 |
together | -0.065 |
there | -0.063 |
Token | sorry |
Token ID | 7926 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
hotel | +0.016 |
ey | +0.016 |
city | +0.016 |
classroom | +0.016 |
principal | +0.016 |
Neg logit contributions | |
es | -0.014 |
on | -0.013 |
together | -0.012 |
icked | -0.012 |
there | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | -8.27e-03 |
Pos logit contributions | |
icked | +0.033 |
es | +0.033 |
pped | +0.027 |
able | +0.026 |
there | +0.024 |
Neg logit contributions | |
hotel | -0.054 |
classroom | -0.053 |
mirror | -0.052 |
talent | -0.051 |
city | -0.051 |
Token | helped |
Token ID | 4193 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
es | +0.013 |
on | +0.012 |
icked | +0.011 |
together | +0.011 |
there | +0.011 |
Neg logit contributions | |
hotel | -0.018 |
ey | -0.017 |
classroom | -0.017 |
principal | -0.017 |
city | -0.017 |
Token | Mom |
Token ID | 11254 |
Feature activation | -4.02e-03 |
Pos logit contributions | |
es | +0.036 |
icked | +0.035 |
there | +0.031 |
on | +0.030 |
together | +0.029 |
Neg logit contributions | |
hotel | -0.037 |
classroom | -0.037 |
principal | -0.036 |
city | -0.036 |
babys | -0.035 |
Token | fix |
Token ID | 4259 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
icked | +0.006 |
es | +0.006 |
on | +0.005 |
together | +0.005 |
there | +0.005 |
Neg logit contributions | |
hotel | -0.008 |
classroom | -0.008 |
principal | -0.008 |
playground | -0.008 |
city | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
es | +0.011 |
on | +0.009 |
together | +0.009 |
icked | +0.009 |
there | +0.009 |
Neg logit contributions | |
hotel | -0.028 |
ey | -0.028 |
city | -0.028 |
classroom | -0.028 |
principal | -0.027 |
Token | fence |
Token ID | 13990 |
Feature activation | -2.97e-02 |
Pos logit contributions | |
ey | +0.244 |
principal | +0.239 |
actor | +0.237 |
Ta | +0.234 |
priest | +0.231 |
Neg logit contributions | |
on | -0.391 |
there | -0.387 |
together | -0.369 |
in | -0.358 |
es | -0.355 |
Token | and |
Token ID | 290 |
Feature activation | -2.70e-03 |
Pos logit contributions | |
es | +0.029 |
icked | +0.028 |
there | +0.021 |
on | +0.021 |
together | +0.019 |
Neg logit contributions | |
hotel | -0.078 |
city | -0.076 |
classroom | -0.075 |
playground | -0.075 |
museum | -0.075 |
Token | clean |
Token ID | 3424 |
Feature activation | -5.36e-03 |
Pos logit contributions | |
es | +0.003 |
icked | +0.003 |
on | +0.003 |
together | +0.003 |
there | +0.003 |
Neg logit contributions | |
hotel | -0.007 |
ey | -0.006 |
classroom | -0.006 |
city | -0.006 |
principal | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
es | +0.006 |
icked | +0.006 |
on | +0.005 |
together | +0.005 |
there | +0.005 |
Neg logit contributions | |
hotel | -0.013 |
city | -0.013 |
classroom | -0.013 |
ey | -0.013 |
playground | -0.013 |
Token | garden |
Token ID | 11376 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
actor | +0.220 |
priest | +0.217 |
principal | +0.215 |
ey | +0.214 |
Ta | +0.212 |
Neg logit contributions | |
on | -0.356 |
there | -0.349 |
together | -0.339 |
in | -0.329 |
es | -0.325 |
Token | she |
Token ID | 673 |
Feature activation | -7.40e-03 |
Pos logit contributions | |
es | +0.026 |
icked | +0.024 |
together | +0.022 |
on | +0.020 |
against | +0.019 |
Neg logit contributions | |
hotel | -0.067 |
priest | -0.065 |
babys | -0.065 |
principal | -0.065 |
museum | -0.063 |
Token | would |
Token ID | 561 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
es | +0.010 |
together | +0.009 |
icked | +0.009 |
on | +0.009 |
there | +0.009 |
Neg logit contributions | |
hotel | -0.017 |
city | -0.016 |
classroom | -0.016 |
principal | -0.016 |
ey | -0.016 |
Token | help |
Token ID | 1037 |
Feature activation | -3.59e-02 |
Pos logit contributions | |
es | +0.046 |
on | +0.043 |
icked | +0.042 |
together | +0.042 |
there | +0.041 |
Neg logit contributions | |
hotel | -0.041 |
ey | -0.040 |
city | -0.039 |
classroom | -0.039 |
principal | -0.039 |
Token | him |
Token ID | 683 |
Feature activation | -6.42e-03 |
Pos logit contributions | |
icked | +0.045 |
es | +0.043 |
on | +0.035 |
together | +0.035 |
there | +0.035 |
Neg logit contributions | |
hotel | -0.084 |
principal | -0.083 |
babys | -0.083 |
city | -0.081 |
classroom | -0.081 |
Token | fix |
Token ID | 4259 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
es | +0.012 |
icked | +0.011 |
on | +0.010 |
together | +0.010 |
there | +0.010 |
Neg logit contributions | |
hotel | -0.011 |
playground | -0.011 |
classroom | -0.011 |
city | -0.011 |
beach | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
es | +0.021 |
on | +0.018 |
icked | +0.018 |
together | +0.018 |
there | +0.017 |
Neg logit contributions | |
hotel | -0.054 |
ey | -0.053 |
city | -0.052 |
classroom | -0.052 |
principal | -0.052 |
Token | fence |
Token ID | 13990 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
ey | +0.238 |
principal | +0.226 |
hotel | +0.219 |
actor | +0.215 |
priest | +0.215 |
Neg logit contributions | |
on | -0.408 |
there | -0.398 |
together | -0.391 |
es | -0.385 |
in | -0.374 |
Token | and |
Token ID | 290 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
es | +0.035 |
icked | +0.034 |
on | +0.025 |
there | +0.024 |
together | +0.023 |
Neg logit contributions | |
hotel | -0.100 |
city | -0.098 |
playground | -0.097 |
classroom | -0.096 |
museum | -0.096 |
Token | told |
Token ID | 1297 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
es | +0.024 |
on | +0.022 |
together | +0.021 |
icked | +0.021 |
there | +0.021 |
Neg logit contributions | |
hotel | -0.030 |
ey | -0.029 |
city | -0.028 |
classroom | -0.028 |
principal | -0.028 |
Token | him |
Token ID | 683 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
es | +0.028 |
on | +0.025 |
icked | +0.024 |
together | +0.024 |
there | +0.023 |
Neg logit contributions | |
hotel | -0.058 |
ey | -0.056 |
city | -0.056 |
classroom | -0.055 |
principal | -0.055 |
Token | to |
Token ID | 284 |
Feature activation | -7.09e-03 |
Pos logit contributions | |
es | +0.012 |
icked | +0.011 |
together | +0.010 |
on | +0.010 |
against | +0.009 |
Neg logit contributions | |
hotel | -0.030 |
principal | -0.029 |
museum | -0.028 |
classroom | -0.028 |
city | -0.028 |
Token | not |
Token ID | 407 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
es | +0.043 |
on | +0.037 |
icked | +0.036 |
together | +0.036 |
there | +0.035 |
Neg logit contributions | |
hotel | -0.105 |
ey | -0.103 |
city | -0.102 |
classroom | -0.102 |
principal | -0.102 |
Token | know |
Token ID | 760 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
hotel | +0.020 |
ey | +0.020 |
city | +0.019 |
classroom | +0.019 |
principal | +0.019 |
Neg logit contributions | |
es | -0.018 |
on | -0.017 |
together | -0.016 |
icked | -0.016 |
there | -0.016 |
Token | how |
Token ID | 703 |
Feature activation | +4.38e-03 |
Pos logit contributions | |
es | +0.028 |
icked | +0.026 |
together | +0.023 |
on | +0.022 |
there | +0.020 |
Neg logit contributions | |
hotel | -0.072 |
principal | -0.070 |
city | -0.069 |
museum | -0.069 |
classroom | -0.069 |
Token | to |
Token ID | 284 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
classroom | +0.011 |
hotel | +0.011 |
principal | +0.011 |
city | +0.011 |
museum | +0.010 |
Neg logit contributions | |
es | -0.005 |
icked | -0.004 |
together | -0.004 |
on | -0.004 |
there | -0.003 |
Token | fix |
Token ID | 4259 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
hotel | +0.044 |
ey | +0.043 |
city | +0.042 |
classroom | +0.041 |
principal | +0.041 |
Neg logit contributions | |
es | -0.039 |
on | -0.037 |
together | -0.036 |
there | -0.036 |
icked | -0.035 |
Token | the |
Token ID | 262 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
es | +0.017 |
icked | +0.015 |
on | +0.014 |
together | +0.014 |
there | +0.014 |
Neg logit contributions | |
hotel | -0.040 |
ey | -0.039 |
city | -0.039 |
classroom | -0.039 |
principal | -0.038 |
Token | computer |
Token ID | 3644 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
majesty | +0.285 |
former | +0.282 |
priest | +0.278 |
Majesty | +0.271 |
hotel | +0.265 |
Neg logit contributions | |
on | -0.382 |
there | -0.342 |
at | -0.334 |
together | -0.326 |
in | -0.324 |
Token | . |
Token ID | 13 |
Feature activation | +1.11e-03 |
Pos logit contributions | |
es | +0.021 |
icked | +0.019 |
on | +0.016 |
there | +0.016 |
together | +0.015 |
Neg logit contributions | |
hotel | -0.069 |
classroom | -0.067 |
city | -0.067 |
principal | -0.067 |
museum | -0.066 |
Token | She |
Token ID | 1375 |
Feature activation | -5.40e-03 |
Pos logit contributions | |
hotel | +0.003 |
ey | +0.003 |
classroom | +0.003 |
city | +0.003 |
principal | +0.003 |
Neg logit contributions | |
es | -0.001 |
on | -0.001 |
icked | -0.001 |
together | -0.001 |
there | -0.001 |
Token | said |
Token ID | 531 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
es | +0.006 |
on | +0.005 |
icked | +0.005 |
together | +0.005 |
there | +0.005 |
Neg logit contributions | |
hotel | -0.014 |
ey | -0.014 |
city | -0.014 |
classroom | -0.014 |
museum | -0.014 |
Token | " |
Token ID | 366 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
es | +0.021 |
icked | +0.018 |
together | +0.018 |
on | +0.017 |
there | +0.016 |
Neg logit contributions | |
principal | -0.047 |
hotel | -0.046 |
classroom | -0.045 |
ey | -0.045 |
priest | -0.045 |
Token | can |
Token ID | 460 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
es | +0.037 |
on | +0.033 |
icked | +0.033 |
together | +0.033 |
there | +0.032 |
Neg logit contributions | |
hotel | -0.057 |
ey | -0.056 |
city | -0.055 |
classroom | -0.055 |
principal | -0.055 |
Token | you |
Token ID | 345 |
Feature activation | +2.10e-02 |
Pos logit contributions | |
es | +0.021 |
icked | +0.019 |
on | +0.019 |
together | +0.019 |
there | +0.017 |
Neg logit contributions | |
hotel | -0.034 |
ey | -0.033 |
principal | -0.033 |
city | -0.032 |
classroom | -0.032 |
Token | help |
Token ID | 1037 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
hotel | +0.035 |
ey | +0.033 |
city | +0.033 |
classroom | +0.033 |
principal | +0.033 |
Neg logit contributions | |
es | -0.043 |
icked | -0.039 |
on | -0.039 |
together | -0.039 |
there | -0.038 |
Token | me |
Token ID | 502 |
Feature activation | +8.17e-03 |
Pos logit contributions | |
es | +0.033 |
icked | +0.031 |
on | +0.029 |
together | +0.029 |
there | +0.029 |
Neg logit contributions | |
hotel | -0.023 |
classroom | -0.022 |
ey | -0.022 |
city | -0.022 |
principal | -0.022 |
Token | restore |
Token ID | 11169 |
Feature activation | -8.88e-03 |
Pos logit contributions | |
hotel | +0.015 |
classroom | +0.015 |
beach | +0.015 |
city | +0.015 |
playground | +0.015 |
Neg logit contributions | |
es | -0.015 |
icked | -0.014 |
together | -0.012 |
on | -0.012 |
there | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
es | +0.012 |
icked | +0.011 |
on | +0.011 |
together | +0.011 |
there | +0.010 |
Neg logit contributions | |
hotel | -0.020 |
ey | -0.020 |
city | -0.019 |
classroom | -0.019 |
principal | -0.019 |
Token | river |
Token ID | 7850 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
principal | +0.242 |
ey | +0.239 |
Ta | +0.235 |
orchestra | +0.226 |
classroom | +0.221 |
Neg logit contributions | |
on | -0.403 |
there | -0.398 |
in | -0.375 |
together | -0.364 |
instead | -0.349 |
Token | ? |
Token ID | 30 |
Feature activation | -5.33e-03 |
Pos logit contributions | |
es | +0.035 |
icked | +0.030 |
on | +0.028 |
together | +0.028 |
there | +0.027 |
Neg logit contributions | |
hotel | -0.075 |
city | -0.073 |
ey | -0.073 |
classroom | -0.073 |
museum | -0.072 |
Token | I |
Token ID | 314 |
Feature activation | -1.47e-03 |
Pos logit contributions | |
es | +0.008 |
icked | +0.007 |
on | +0.007 |
together | +0.007 |
there | +0.007 |
Neg logit contributions | |
ey | -0.011 |
museum | -0.011 |
hotel | -0.011 |
principal | -0.011 |
city | -0.011 |
Token | can |
Token ID | 460 |
Feature activation | -3.04e-04 |
Pos logit contributions | |
es | +0.002 |
on | +0.002 |
together | +0.002 |
there | +0.002 |
icked | +0.002 |
Neg logit contributions | |
hotel | -0.003 |
ey | -0.003 |
principal | -0.003 |
city | -0.003 |
classroom | -0.003 |
Token | 't |
Token ID | 470 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
es | +0.001 |
on | +0.001 |
icked | +0.001 |
together | +0.001 |
there | +0.001 |
Neg logit contributions | |
hotel | -0.001 |
ey | -0.001 |
city | -0.001 |
classroom | -0.001 |
principal | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | -4.39e-03 |
Pos logit contributions | |
es | +0.001 |
on | +0.001 |
together | +0.001 |
icked | +0.001 |
there | +0.001 |
Neg logit contributions | |
hotel | -0.002 |
ey | -0.002 |
city | -0.002 |
classroom | -0.002 |
principal | -0.002 |
Token | " |
Token ID | 1 |
Feature activation | +1.72e-03 |
Pos logit contributions | |
es | +0.004 |
on | +0.004 |
together | +0.004 |
there | +0.003 |
icked | +0.003 |
Neg logit contributions | |
hotel | -0.012 |
ey | -0.012 |
city | -0.012 |
principal | -0.012 |
classroom | -0.012 |
Token | Can |
Token ID | 6090 |
Feature activation | -3.40e-02 |
Pos logit contributions | |
hotel | +0.004 |
ey | +0.004 |
city | +0.004 |
classroom | +0.004 |
doorway | +0.004 |
Neg logit contributions | |
together | -0.002 |
es | -0.002 |
on | -0.002 |
there | -0.002 |
instead | -0.002 |
Token | we |
Token ID | 356 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
es | +0.036 |
on | +0.031 |
icked | +0.030 |
together | +0.030 |
there | +0.029 |
Neg logit contributions | |
hotel | -0.090 |
ey | -0.088 |
city | -0.087 |
classroom | -0.087 |
principal | -0.087 |
Token | fix |
Token ID | 4259 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
hotel | +0.040 |
ey | +0.039 |
principal | +0.038 |
city | +0.038 |
classroom | +0.038 |
Neg logit contributions | |
es | -0.040 |
on | -0.038 |
together | -0.037 |
there | -0.037 |
icked | -0.037 |
Token | the |
Token ID | 262 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
es | +0.027 |
on | +0.024 |
together | +0.023 |
icked | +0.023 |
there | +0.023 |
Neg logit contributions | |
hotel | -0.053 |
ey | -0.052 |
city | -0.051 |
classroom | -0.051 |
principal | -0.051 |
Token | blanket |
Token ID | 18447 |
Feature activation | -4.26e-02 |
Pos logit contributions | |
principal | +0.317 |
ey | +0.308 |
courtroom | +0.296 |
carn | +0.291 |
hotel | +0.291 |
Neg logit contributions | |
on | -0.330 |
there | -0.321 |
es | -0.308 |
together | -0.303 |
in | -0.291 |
Token | , |
Token ID | 11 |
Feature activation | -4.49e-03 |
Pos logit contributions | |
es | +0.057 |
icked | +0.053 |
on | +0.039 |
together | +0.038 |
ged | +0.038 |
Neg logit contributions | |
hotel | -0.102 |
classroom | -0.098 |
city | -0.097 |
salon | -0.095 |
playground | -0.095 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
es | +0.008 |
on | +0.008 |
together | +0.008 |
there | +0.007 |
icked | +0.007 |
Neg logit contributions | |
hotel | -0.008 |
ey | -0.008 |
city | -0.008 |
classroom | -0.008 |
principal | -0.008 |
Token | my |
Token ID | 1820 |
Feature activation | -3.31e-02 |
Pos logit contributions | |
es | +0.039 |
icked | +0.038 |
against | +0.032 |
on | +0.032 |
bed | +0.032 |
Neg logit contributions | |
classroom | -0.028 |
hotel | -0.027 |
city | -0.026 |
principal | -0.026 |
museum | -0.026 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
icked | +0.064 |
es | +0.063 |
bed | +0.051 |
ged | +0.049 |
bled | +0.048 |
Neg logit contributions | |
classroom | -0.056 |
hotel | -0.056 |
principal | -0.055 |
airport | -0.055 |
museum | -0.055 |
Token | person |
Token ID | 1048 |
Feature activation | -2.30e-02 |
Pos logit contributions | |
opl | +0.291 |
ey | +0.269 |
Ta | +0.255 |
carn | +0.245 |
former | +0.237 |
Neg logit contributions | |
there | -0.282 |
in | -0.281 |
on | -0.271 |
with | -0.256 |
near | -0.234 |
Token | used |
Token ID | 973 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
es | +0.032 |
on | +0.029 |
together | +0.028 |
icked | +0.028 |
there | +0.027 |
Neg logit contributions | |
hotel | -0.053 |
ey | -0.052 |
city | -0.051 |
classroom | -0.051 |
principal | -0.051 |
Token | magic |
Token ID | 5536 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
es | +0.036 |
icked | +0.030 |
on | +0.030 |
together | +0.029 |
there | +0.028 |
Neg logit contributions | |
hotel | -0.087 |
ey | -0.085 |
city | -0.084 |
classroom | -0.084 |
principal | -0.083 |
Token | to |
Token ID | 284 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
hotel | +0.037 |
ey | +0.036 |
city | +0.035 |
classroom | +0.035 |
principal | +0.035 |
Neg logit contributions | |
es | -0.022 |
on | -0.020 |
icked | -0.020 |
together | -0.020 |
there | -0.019 |
Token | restore |
Token ID | 11169 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
hotel | +0.023 |
ey | +0.022 |
principal | +0.022 |
classroom | +0.022 |
city | +0.022 |
Neg logit contributions | |
es | -0.019 |
on | -0.018 |
there | -0.018 |
together | -0.018 |
icked | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
es | +0.021 |
on | +0.018 |
icked | +0.018 |
together | +0.018 |
there | +0.017 |
Neg logit contributions | |
hotel | -0.056 |
ey | -0.055 |
city | -0.054 |
classroom | -0.054 |
principal | -0.054 |
Token | poor |
Token ID | 3595 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
principal | +0.326 |
carn | +0.324 |
Ta | +0.314 |
opl | +0.297 |
actor | +0.293 |
Neg logit contributions | |
there | -0.337 |
on | -0.335 |
in | -0.330 |
together | -0.308 |
with | -0.305 |
Token | bulb |
Token ID | 28287 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
Ta | +0.197 |
carn | +0.191 |
principal | +0.191 |
ey | +0.190 |
hotel | +0.183 |
Neg logit contributions | |
there | -0.278 |
on | -0.273 |
es | -0.271 |
in | -0.263 |
together | -0.260 |
Token | . |
Token ID | 13 |
Feature activation | -7.88e-04 |
Pos logit contributions | |
es | +0.027 |
icked | +0.026 |
there | +0.020 |
on | +0.020 |
together | +0.019 |
Neg logit contributions | |
hotel | -0.082 |
city | -0.080 |
classroom | -0.079 |
ey | -0.079 |
playground | -0.078 |
Token | They |
Token ID | 1119 |
Feature activation | -2.45e-03 |
Pos logit contributions | |
es | +0.001 |
on | +0.001 |
icked | +0.001 |
there | +0.001 |
together | +0.001 |
Neg logit contributions | |
hotel | -0.002 |
ey | -0.002 |
city | -0.002 |
classroom | -0.002 |
museum | -0.002 |
Token | brushed |
Token ID | 31782 |
Feature activation | -4.72e-02 |
Pos logit contributions | |
es | +0.004 |
on | +0.003 |
icked | +0.003 |
together | +0.003 |
there | +0.003 |
Neg logit contributions | |
hotel | -0.005 |
ey | -0.005 |
city | -0.005 |
classroom | -0.005 |
principal | -0.005 |
Token | been |
Token ID | 587 |
Feature activation | +8.74e-03 |
Pos logit contributions | |
es | +0.021 |
icked | +0.019 |
together | +0.018 |
on | +0.018 |
there | +0.017 |
Neg logit contributions | |
classroom | -0.022 |
hotel | -0.022 |
talent | -0.022 |
principal | -0.022 |
playground | -0.022 |
Token | able |
Token ID | 1498 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
hotel | +0.020 |
principal | +0.020 |
classroom | +0.020 |
audience | +0.020 |
talent | +0.019 |
Neg logit contributions | |
es | -0.011 |
icked | -0.010 |
together | -0.008 |
on | -0.008 |
instead | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
es | +0.029 |
icked | +0.024 |
together | +0.020 |
on | +0.019 |
there | +0.018 |
Neg logit contributions | |
hotel | -0.061 |
classroom | -0.059 |
city | -0.058 |
playground | -0.058 |
talent | -0.058 |
Token | help |
Token ID | 1037 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
hotel | +0.045 |
ey | +0.043 |
city | +0.043 |
classroom | +0.042 |
principal | +0.042 |
Neg logit contributions | |
es | -0.048 |
on | -0.045 |
icked | -0.044 |
together | -0.044 |
there | -0.044 |
Token | wrap |
Token ID | 14441 |
Feature activation | -4.66e-02 |
Pos logit contributions | |
es | +0.013 |
icked | +0.012 |
together | +0.011 |
on | +0.010 |
there | +0.010 |
Neg logit contributions | |
hotel | -0.038 |
ey | -0.037 |
classroom | -0.037 |
principal | -0.037 |
city | -0.037 |
Token | the |
Token ID | 262 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
es | +0.050 |
icked | +0.044 |
on | +0.044 |
together | +0.043 |
there | +0.042 |
Neg logit contributions | |
hotel | -0.121 |
ey | -0.118 |
city | -0.116 |
classroom | -0.116 |
principal | -0.116 |
Token | fancy |
Token ID | 14996 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
former | +0.313 |
opl | +0.278 |
igsaw | +0.277 |
salon | +0.259 |
gol | +0.258 |
Neg logit contributions | |
on | -0.362 |
there | -0.356 |
in | -0.323 |
with | -0.305 |
at | -0.296 |
Token | suit |
Token ID | 6050 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
former | +0.226 |
Ta | +0.215 |
igsaw | +0.197 |
principal | +0.195 |
sidelines | +0.195 |
Neg logit contributions | |
on | -0.304 |
there | -0.293 |
with | -0.278 |
in | -0.276 |
at | -0.270 |
Token | for |
Token ID | 329 |
Feature activation | +2.47e-02 |
Pos logit contributions | |
es | +0.030 |
icked | +0.028 |
on | +0.023 |
there | +0.022 |
together | +0.022 |
Neg logit contributions | |
hotel | -0.090 |
classroom | -0.088 |
city | -0.087 |
principal | -0.087 |
playground | -0.087 |
Token | the |
Token ID | 262 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
hotel | +0.058 |
city | +0.056 |
classroom | +0.056 |
principal | +0.056 |
ey | +0.055 |
Neg logit contributions | |
es | -0.033 |
icked | -0.030 |
on | -0.028 |
together | -0.028 |
there | -0.027 |
Token | special |
Token ID | 2041 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
ey | +0.295 |
opl | +0.272 |
former | +0.268 |
igsaw | +0.260 |
majesty | +0.254 |
Neg logit contributions | |
on | -0.244 |
there | -0.242 |
in | -0.225 |
with | -0.207 |
it | -0.195 |
Token | why |
Token ID | 1521 |
Feature activation | -5.96e-05 |
Pos logit contributions | |
es | +0.036 |
icked | +0.034 |
together | +0.028 |
on | +0.027 |
there | +0.025 |
Neg logit contributions | |
hotel | -0.106 |
museum | -0.103 |
principal | -0.102 |
classroom | -0.102 |
city | -0.101 |
Token | he |
Token ID | 339 |
Feature activation | -7.77e-03 |
Pos logit contributions | |
es | +0.000 |
icked | +0.000 |
together | +0.000 |
on | +0.000 |
against | +0.000 |
Neg logit contributions | |
hotel | -0.000 |
classroom | -0.000 |
city | -0.000 |
museum | -0.000 |
principal | -0.000 |
Token | had |
Token ID | 550 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
es | +0.012 |
together | +0.011 |
icked | +0.011 |
on | +0.011 |
there | +0.010 |
Neg logit contributions | |
hotel | -0.016 |
classroom | -0.016 |
city | -0.016 |
museum | -0.016 |
ey | -0.016 |
Token | fixed |
Token ID | 5969 |
Feature activation | -2.77e-02 |
Pos logit contributions | |
es | +0.027 |
on | +0.025 |
there | +0.024 |
together | +0.024 |
icked | +0.024 |
Neg logit contributions | |
hotel | -0.053 |
ey | -0.052 |
city | -0.051 |
classroom | -0.051 |
principal | -0.051 |
Token | her |
Token ID | 607 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
es | +0.025 |
icked | +0.023 |
together | +0.020 |
on | +0.020 |
there | +0.019 |
Neg logit contributions | |
hotel | -0.076 |
ey | -0.074 |
classroom | -0.074 |
city | -0.074 |
principal | -0.073 |
Token | the |
Token ID | 262 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
hotel | +0.151 |
ey | +0.149 |
principal | +0.146 |
city | +0.142 |
former | +0.139 |
Neg logit contributions | |
es | -0.281 |
on | -0.273 |
there | -0.265 |
together | -0.264 |
icked | -0.253 |
Token | screen |
Token ID | 3159 |
Feature activation | -3.87e-02 |
Pos logit contributions | |
former | +0.254 |
ey | +0.245 |
salon | +0.241 |
majesty | +0.238 |
principal | +0.238 |
Neg logit contributions | |
on | -0.402 |
there | -0.366 |
in | -0.363 |
together | -0.362 |
at | -0.352 |
Token | . |
Token ID | 13 |
Feature activation | -7.99e-03 |
Pos logit contributions | |
es | +0.038 |
icked | +0.037 |
there | +0.026 |
on | +0.026 |
together | +0.026 |
Neg logit contributions | |
hotel | -0.103 |
classroom | -0.103 |
city | -0.101 |
museum | -0.100 |
playground | -0.100 |
Token |
|
Token ID | 198 |
Feature activation | -3.30e-03 |
Pos logit contributions | |
es | +0.013 |
on | +0.012 |
together | +0.012 |
icked | +0.012 |
there | +0.012 |
Neg logit contributions | |
hotel | -0.016 |
ey | -0.016 |
city | -0.016 |
principal | -0.015 |
classroom | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | -7.77e-03 |
Pos logit contributions | |
es | +0.004 |
on | +0.003 |
together | +0.003 |
there | +0.003 |
icked | +0.003 |
Neg logit contributions | |
hotel | -0.009 |
ey | -0.008 |
city | -0.008 |
principal | -0.008 |
classroom | -0.008 |
Token | " |
Token ID | 1 |
Feature activation | -6.59e-03 |
Pos logit contributions | |
es | +0.013 |
on | +0.012 |
together | +0.012 |
there | +0.011 |
icked | +0.011 |
Neg logit contributions | |
hotel | -0.016 |
ey | -0.016 |
city | -0.015 |
principal | -0.015 |
classroom | -0.015 |
Token | you |
Token ID | 345 |
Feature activation | +1.40e-02 |
Pos logit contributions | |
es | +0.016 |
icked | +0.015 |
on | +0.014 |
together | +0.013 |
there | +0.013 |
Neg logit contributions | |
hotel | -0.027 |
ey | -0.026 |
classroom | -0.026 |
city | -0.026 |
museum | -0.026 |
Token | will |
Token ID | 481 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
hotel | +0.032 |
ey | +0.031 |
classroom | +0.031 |
city | +0.030 |
principal | +0.030 |
Neg logit contributions | |
es | -0.021 |
on | -0.019 |
together | -0.019 |
icked | -0.018 |
there | -0.018 |
Token | have |
Token ID | 423 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
on | +0.022 |
there | +0.021 |
together | +0.021 |
es | +0.021 |
instead | +0.020 |
Neg logit contributions | |
ey | -0.036 |
hotel | -0.035 |
principal | -0.034 |
city | -0.034 |
doorway | -0.033 |
Token | to |
Token ID | 284 |
Feature activation | +6.58e-03 |
Pos logit contributions | |
es | +0.016 |
icked | +0.013 |
on | +0.013 |
together | +0.012 |
there | +0.012 |
Neg logit contributions | |
hotel | -0.038 |
principal | -0.037 |
classroom | -0.037 |
city | -0.037 |
ey | -0.037 |
Token | fix |
Token ID | 4259 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
hotel | +0.015 |
ey | +0.014 |
doorway | +0.014 |
city | +0.014 |
principal | +0.014 |
Neg logit contributions | |
on | -0.009 |
together | -0.009 |
with | -0.009 |
there | -0.009 |
in | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
es | +0.030 |
on | +0.026 |
together | +0.025 |
icked | +0.024 |
there | +0.024 |
Neg logit contributions | |
hotel | -0.081 |
ey | -0.080 |
city | -0.079 |
classroom | -0.079 |
principal | -0.079 |
Token | veil |
Token ID | 30615 |
Feature activation | -4.06e-02 |
Pos logit contributions | |
former | +0.290 |
igsaw | +0.289 |
principal | +0.280 |
gol | +0.277 |
carn | +0.274 |
Neg logit contributions | |
on | -0.368 |
there | -0.360 |
in | -0.326 |
it | -0.317 |
at | -0.307 |
Token | with |
Token ID | 351 |
Feature activation | +8.61e-03 |
Pos logit contributions | |
icked | +0.039 |
es | +0.037 |
on | +0.028 |
there | +0.027 |
together | +0.026 |
Neg logit contributions | |
hotel | -0.108 |
city | -0.104 |
classroom | -0.104 |
playground | -0.102 |
principal | -0.102 |
Token | some |
Token ID | 617 |
Feature activation | +4.13e-02 |
Pos logit contributions | |
hotel | +0.019 |
ey | +0.019 |
city | +0.019 |
principal | +0.019 |
classroom | +0.019 |
Neg logit contributions | |
es | -0.012 |
on | -0.011 |
together | -0.011 |
there | -0.011 |
icked | -0.011 |
Token | thread |
Token ID | 4704 |
Feature activation | -5.12e-02 |
Pos logit contributions | |
hotel | +0.075 |
principal | +0.073 |
ey | +0.073 |
classroom | +0.072 |
city | +0.072 |
Neg logit contributions | |
es | -0.076 |
on | -0.075 |
there | -0.071 |
together | -0.071 |
icked | -0.068 |
Token | and |
Token ID | 290 |
Feature activation | -7.80e-03 |
Pos logit contributions | |
es | +0.043 |
icked | +0.041 |
on | +0.034 |
there | +0.034 |
together | +0.032 |
Neg logit contributions | |
hotel | -0.139 |
city | -0.133 |
classroom | -0.133 |
ey | -0.132 |
principal | -0.132 |
Token | ." |
Token ID | 526 |
Feature activation | -9.92e-04 |
Pos logit contributions | |
es | +0.015 |
icked | +0.014 |
on | +0.012 |
there | +0.011 |
together | +0.011 |
Neg logit contributions | |
hotel | -0.042 |
city | -0.041 |
classroom | -0.041 |
museum | -0.040 |
playground | -0.040 |
Token |
|
Token ID | 198 |
Feature activation | +2.86e-03 |
Pos logit contributions | |
es | +0.001 |
icked | +0.001 |
on | +0.001 |
together | +0.001 |
there | +0.001 |
Neg logit contributions | |
hotel | -0.002 |
ey | -0.002 |
principal | -0.002 |
classroom | -0.002 |
city | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -2.65e-03 |
Pos logit contributions | |
hotel | +0.007 |
ey | +0.007 |
city | +0.007 |
classroom | +0.007 |
principal | +0.007 |
Neg logit contributions | |
es | -0.003 |
on | -0.003 |
together | -0.003 |
icked | -0.003 |
there | -0.003 |
Token | " |
Token ID | 1 |
Feature activation | +5.14e-03 |
Pos logit contributions | |
es | +0.004 |
on | +0.004 |
together | +0.004 |
there | +0.004 |
icked | +0.004 |
Neg logit contributions | |
hotel | -0.005 |
ey | -0.005 |
city | -0.005 |
classroom | -0.005 |
principal | -0.005 |
Token | Number |
Token ID | 15057 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
hotel | +0.011 |
ey | +0.010 |
city | +0.010 |
doorway | +0.010 |
classroom | +0.010 |
Neg logit contributions | |
together | -0.008 |
es | -0.008 |
on | -0.008 |
there | -0.008 |
instead | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
hotel | +0.046 |
museum | +0.045 |
city | +0.045 |
classroom | +0.044 |
principal | +0.043 |
Neg logit contributions | |
es | -0.024 |
icked | -0.023 |
together | -0.021 |
on | -0.021 |
against | -0.020 |
Token | earth |
Token ID | 4534 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
hotel | +0.228 |
principal | +0.226 |
salon | +0.219 |
babys | +0.214 |
carn | +0.211 |
Neg logit contributions | |
on | -0.413 |
there | -0.398 |
in | -0.390 |
together | -0.390 |
es | -0.374 |
Token | ? |
Token ID | 30 |
Feature activation | -6.22e-03 |
Pos logit contributions | |
es | +0.030 |
icked | +0.028 |
on | +0.027 |
together | +0.027 |
there | +0.027 |
Neg logit contributions | |
hotel | -0.034 |
ey | -0.032 |
city | -0.032 |
classroom | -0.032 |
museum | -0.032 |
Token | What |
Token ID | 1867 |
Feature activation | +3.32e-03 |
Pos logit contributions | |
es | +0.010 |
icked | +0.009 |
on | +0.009 |
together | +0.008 |
there | +0.008 |
Neg logit contributions | |
hotel | -0.013 |
principal | -0.013 |
city | -0.013 |
museum | -0.013 |
classroom | -0.013 |
Token | does |
Token ID | 857 |
Feature activation | +1.76e-03 |
Pos logit contributions | |
museum | +0.006 |
hotel | +0.006 |
classroom | +0.006 |
city | +0.006 |
principal | +0.006 |
Neg logit contributions | |
es | -0.006 |
icked | -0.006 |
together | -0.005 |
on | -0.005 |
against | -0.005 |
Token | that |
Token ID | 326 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
hotel | +0.004 |
museum | +0.004 |
city | +0.004 |
classroom | +0.004 |
talent | +0.004 |
Neg logit contributions | |
es | -0.003 |
icked | -0.003 |
together | -0.002 |
on | -0.002 |
ed | -0.002 |
Token | they |
Token ID | 484 |
Feature activation | -4.53e-04 |
Pos logit contributions | |
es | +0.004 |
icked | +0.004 |
on | +0.003 |
together | +0.003 |
against | +0.003 |
Neg logit contributions | |
hotel | -0.011 |
principal | -0.011 |
classroom | -0.011 |
city | -0.011 |
priest | -0.011 |
Token | have |
Token ID | 423 |
Feature activation | -5.54e-03 |
Pos logit contributions | |
on | +0.001 |
es | +0.001 |
together | +0.001 |
there | +0.001 |
in | +0.001 |
Neg logit contributions | |
ey | -0.001 |
hotel | -0.001 |
classroom | -0.001 |
principal | -0.001 |
doorway | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | +4.37e-03 |
Pos logit contributions | |
es | +0.006 |
on | +0.005 |
icked | +0.005 |
together | +0.004 |
there | +0.004 |
Neg logit contributions | |
hotel | -0.015 |
city | -0.015 |
ey | -0.015 |
classroom | -0.015 |
principal | -0.015 |
Token | help |
Token ID | 1037 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
ey | +0.010 |
hotel | +0.009 |
doorway | +0.009 |
principal | +0.009 |
museum | +0.009 |
Neg logit contributions | |
on | -0.006 |
together | -0.006 |
there | -0.006 |
instead | -0.006 |
with | -0.006 |
Token | fix |
Token ID | 4259 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
es | +0.033 |
icked | +0.032 |
there | +0.027 |
on | +0.027 |
together | +0.026 |
Neg logit contributions | |
hotel | -0.048 |
city | -0.046 |
principal | -0.046 |
classroom | -0.046 |
babys | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
es | +0.019 |
icked | +0.016 |
on | +0.016 |
together | +0.016 |
there | +0.015 |
Neg logit contributions | |
hotel | -0.049 |
ey | -0.048 |
city | -0.048 |
classroom | -0.047 |
principal | -0.047 |
Token | fence |
Token ID | 13990 |
Feature activation | -3.24e-02 |
Pos logit contributions | |
Ta | +0.276 |
actor | +0.269 |
orchestra | +0.268 |
ey | +0.268 |
priest | +0.263 |
Neg logit contributions | |
on | -0.358 |
there | -0.348 |
at | -0.322 |
in | -0.318 |
with | -0.314 |
Token | . |
Token ID | 13 |
Feature activation | +9.27e-04 |
Pos logit contributions | |
es | +0.029 |
icked | +0.027 |
on | +0.022 |
there | +0.021 |
together | +0.020 |
Neg logit contributions | |
hotel | -0.089 |
city | -0.087 |
classroom | -0.086 |
playground | -0.085 |
museum | -0.085 |
Token |
|
Token ID | 198 |
Feature activation | +3.03e-04 |
Pos logit contributions | |
hotel | +0.002 |
classroom | +0.002 |
city | +0.002 |
ey | +0.002 |
principal | +0.002 |
Neg logit contributions | |
es | -0.001 |
icked | -0.001 |
on | -0.001 |
together | -0.001 |
there | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -7.60e-03 |
Pos logit contributions | |
hotel | +0.001 |
ey | +0.001 |
city | +0.001 |
classroom | +0.001 |
principal | +0.001 |
Neg logit contributions | |
es | -0.000 |
on | -0.000 |
together | -0.000 |
there | -0.000 |
icked | -0.000 |
Token | Tom |
Token ID | 13787 |
Feature activation | +6.15e-03 |
Pos logit contributions | |
es | +0.010 |
together | +0.009 |
on | +0.009 |
there | +0.008 |
icked | +0.008 |
Neg logit contributions | |
hotel | -0.019 |
ey | -0.018 |
city | -0.018 |
classroom | -0.018 |
principal | -0.018 |
Token | tutor |
Token ID | 48680 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
ey | +0.169 |
Ta | +0.147 |
principal | +0.142 |
opl | +0.139 |
orchestra | +0.132 |
Neg logit contributions | |
together | -0.165 |
there | -0.163 |
on | -0.161 |
it | -0.144 |
in | -0.143 |
Token | to |
Token ID | 284 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
es | +0.028 |
icked | +0.027 |
together | +0.021 |
on | +0.021 |
there | +0.020 |
Neg logit contributions | |
hotel | -0.083 |
classroom | -0.081 |
principal | -0.080 |
city | -0.080 |
museum | -0.080 |
Token | help |
Token ID | 1037 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
ey | +0.034 |
hotel | +0.034 |
city | +0.032 |
principal | +0.032 |
classroom | +0.032 |
Neg logit contributions | |
on | -0.030 |
es | -0.030 |
together | -0.029 |
there | -0.029 |
instead | -0.028 |
Token | him |
Token ID | 683 |
Feature activation | +3.33e-03 |
Pos logit contributions | |
es | +0.022 |
icked | +0.021 |
together | +0.018 |
on | +0.018 |
there | +0.018 |
Neg logit contributions | |
hotel | -0.055 |
principal | -0.053 |
classroom | -0.053 |
city | -0.053 |
museum | -0.052 |
Token | fix |
Token ID | 4259 |
Feature activation | -9.09e-03 |
Pos logit contributions | |
hotel | +0.008 |
classroom | +0.008 |
playground | +0.008 |
city | +0.008 |
beach | +0.008 |
Neg logit contributions | |
es | -0.004 |
icked | -0.004 |
on | -0.003 |
there | -0.003 |
together | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
es | +0.010 |
on | +0.008 |
icked | +0.008 |
together | +0.008 |
there | +0.008 |
Neg logit contributions | |
hotel | -0.024 |
ey | -0.023 |
city | -0.023 |
classroom | -0.023 |
principal | -0.023 |
Token | boat |
Token ID | 8848 |
Feature activation | -2.78e-02 |
Pos logit contributions | |
ey | +0.269 |
principal | +0.253 |
Ta | +0.235 |
babys | +0.228 |
classroom | +0.227 |
Neg logit contributions | |
on | -0.387 |
there | -0.382 |
together | -0.363 |
in | -0.356 |
it | -0.336 |
Token | . |
Token ID | 13 |
Feature activation | -1.54e-03 |
Pos logit contributions | |
es | +0.025 |
icked | +0.024 |
on | +0.018 |
there | +0.018 |
together | +0.017 |
Neg logit contributions | |
hotel | -0.077 |
city | -0.074 |
playground | -0.074 |
classroom | -0.073 |
museum | -0.073 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
es | +0.002 |
on | +0.002 |
together | +0.002 |
icked | +0.002 |
there | +0.002 |
Neg logit contributions | |
hotel | -0.004 |
ey | -0.004 |
city | -0.004 |
classroom | -0.004 |
principal | -0.004 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.24e-02 |
Pos logit contributions | |
hotel | +0.025 |
city | +0.025 |
museum | +0.024 |
classroom | +0.024 |
playground | +0.024 |
Neg logit contributions | |
icked | -0.027 |
es | -0.026 |
able | -0.025 |
as | -0.024 |
instead | -0.024 |
Token | upon |
Token ID | 2402 |
Feature activation | -3.32e-03 |
Pos logit contributions | |
hotel | +0.057 |
classroom | +0.053 |
ey | +0.053 |
city | +0.053 |
principal | +0.052 |
Neg logit contributions | |
es | -0.064 |
icked | -0.058 |
on | -0.058 |
together | -0.057 |
there | -0.055 |
Token |
|
Token ID | 198 |
Feature activation | -2.59e-03 |
Pos logit contributions | |
hotel | +0.007 |
ey | +0.007 |
classroom | +0.007 |
city | +0.007 |
principal | +0.006 |
Neg logit contributions | |
es | -0.004 |
on | -0.003 |
icked | -0.003 |
together | -0.003 |
there | -0.003 |
Token | " |
Token ID | 1 |
Feature activation | +2.18e-03 |
Pos logit contributions | |
es | +0.004 |
icked | +0.003 |
on | +0.003 |
there | +0.003 |
together | +0.003 |
Neg logit contributions | |
hotel | -0.006 |
classroom | -0.006 |
city | -0.006 |
ey | -0.006 |
principal | -0.006 |
Token | Can |
Token ID | 6090 |
Feature activation | -3.33e-02 |
Pos logit contributions | |
hotel | +0.004 |
ey | +0.004 |
city | +0.004 |
classroom | +0.004 |
principal | +0.004 |
Neg logit contributions | |
es | -0.004 |
together | -0.003 |
on | -0.003 |
there | -0.003 |
icked | -0.003 |
Token | we |
Token ID | 356 |
Feature activation | +2.20e-02 |
Pos logit contributions | |
es | +0.041 |
icked | +0.036 |
on | +0.034 |
together | +0.033 |
there | +0.031 |
Neg logit contributions | |
hotel | -0.082 |
principal | -0.080 |
classroom | -0.080 |
city | -0.080 |
ey | -0.079 |
Token | repair |
Token ID | 9185 |
Feature activation | -3.03e-02 |
Pos logit contributions | |
hotel | +0.041 |
ey | +0.040 |
principal | +0.039 |
city | +0.039 |
classroom | +0.039 |
Neg logit contributions | |
es | -0.040 |
on | -0.038 |
together | -0.037 |
there | -0.037 |
icked | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
es | +0.037 |
icked | +0.034 |
on | +0.032 |
together | +0.031 |
there | +0.031 |
Neg logit contributions | |
hotel | -0.074 |
ey | -0.072 |
city | -0.072 |
classroom | -0.071 |
principal | -0.071 |
Token | calendar |
Token ID | 11845 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
former | +0.219 |
Ta | +0.214 |
ey | +0.211 |
carn | +0.208 |
principal | +0.198 |
Neg logit contributions | |
on | -0.435 |
there | -0.412 |
in | -0.394 |
together | -0.387 |
at | -0.380 |
Token | , |
Token ID | 11 |
Feature activation | +2.48e-03 |
Pos logit contributions | |
es | +0.013 |
icked | +0.012 |
on | +0.009 |
there | +0.009 |
together | +0.009 |
Neg logit contributions | |
hotel | -0.031 |
classroom | -0.030 |
city | -0.030 |
museum | -0.029 |
playground | -0.029 |
Token | Mom |
Token ID | 11254 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
hotel | +0.005 |
ey | +0.005 |
city | +0.005 |
principal | +0.005 |
classroom | +0.005 |
Neg logit contributions | |
es | -0.004 |
on | -0.004 |
together | -0.004 |
icked | -0.004 |
there | -0.004 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
es | +0.044 |
icked | +0.043 |
bed | +0.034 |
ged | +0.034 |
against | +0.034 |
Neg logit contributions | |
classroom | -0.049 |
hotel | -0.049 |
city | -0.047 |
museum | -0.047 |
principal | -0.047 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.65e-02 |
Pos logit contributions | |
es | +0.015 |
icked | +0.013 |
on | +0.012 |
together | +0.012 |
there | +0.011 |
Neg logit contributions | |
hotel | -0.026 |
classroom | -0.025 |
city | -0.025 |
principal | -0.025 |
ey | -0.025 |
Token | I |
Token ID | 40 |
Feature activation | +3.28e-03 |
Pos logit contributions | |
hotel | +0.002 |
ey | +0.002 |
city | +0.002 |
classroom | +0.002 |
principal | +0.002 |
Neg logit contributions | |
es | -0.001 |
on | -0.001 |
icked | -0.001 |
together | -0.001 |
there | -0.001 |
Token | can |
Token ID | 460 |
Feature activation | -8.68e-03 |
Pos logit contributions | |
hotel | +0.007 |
ey | +0.006 |
principal | +0.006 |
city | +0.006 |
classroom | +0.006 |
Neg logit contributions | |
es | -0.006 |
on | -0.005 |
together | -0.005 |
there | -0.005 |
icked | -0.005 |
Token | help |
Token ID | 1037 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
es | +0.017 |
icked | +0.016 |
on | +0.016 |
together | +0.015 |
there | +0.015 |
Neg logit contributions | |
hotel | -0.014 |
city | -0.014 |
classroom | -0.014 |
museum | -0.014 |
playground | -0.013 |
Token | you |
Token ID | 345 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
icked | +0.043 |
es | +0.041 |
on | +0.036 |
together | +0.034 |
there | +0.034 |
Neg logit contributions | |
hotel | -0.051 |
museum | -0.050 |
principal | -0.050 |
city | -0.050 |
classroom | -0.049 |
Token | separate |
Token ID | 4553 |
Feature activation | +5.20e-03 |
Pos logit contributions | |
hotel | +0.024 |
classroom | +0.023 |
playground | +0.023 |
city | +0.023 |
museum | +0.023 |
Neg logit contributions | |
es | -0.021 |
icked | -0.021 |
on | -0.017 |
led | -0.016 |
ged | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
hotel | +0.012 |
ey | +0.012 |
principal | +0.012 |
classroom | +0.012 |
city | +0.012 |
Neg logit contributions | |
es | -0.007 |
on | -0.007 |
together | -0.006 |
there | -0.006 |
icked | -0.006 |
Token | toys |
Token ID | 14958 |
Feature activation | -3.43e-02 |
Pos logit contributions | |
former | +0.359 |
Ta | +0.354 |
opl | +0.351 |
orchestra | +0.339 |
principal | +0.336 |
Neg logit contributions | |
there | -0.299 |
on | -0.291 |
in | -0.270 |
with | -0.248 |
es | -0.245 |
Token | ." |
Token ID | 526 |
Feature activation | -5.24e-03 |
Pos logit contributions | |
icked | +0.042 |
es | +0.041 |
ged | +0.030 |
on | +0.029 |
against | +0.028 |
Neg logit contributions | |
hotel | -0.082 |
city | -0.079 |
museum | -0.079 |
beach | -0.078 |
classroom | -0.078 |
Token |
|
Token ID | 198 |
Feature activation | -2.49e-03 |
Pos logit contributions | |
es | +0.008 |
icked | +0.008 |
on | +0.007 |
there | +0.007 |
together | +0.007 |
Neg logit contributions | |
hotel | -0.011 |
city | -0.011 |
museum | -0.011 |
principal | -0.011 |
classroom | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | -6.93e-03 |
Pos logit contributions | |
es | +0.004 |
icked | +0.004 |
on | +0.003 |
together | +0.003 |
there | +0.003 |
Neg logit contributions | |
hotel | -0.005 |
city | -0.005 |
museum | -0.005 |
classroom | -0.005 |
principal | -0.005 |
Token | So |
Token ID | 2396 |
Feature activation | -2.93e-02 |
Pos logit contributions | |
es | +0.011 |
on | +0.010 |
icked | +0.010 |
together | +0.010 |
there | +0.010 |
Neg logit contributions | |
hotel | -0.014 |
ey | -0.014 |
city | -0.014 |
classroom | -0.014 |
principal | -0.014 |
Token | can |
Token ID | 460 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
es | +0.032 |
icked | +0.029 |
on | +0.028 |
together | +0.028 |
there | +0.027 |
Neg logit contributions | |
hotel | -0.049 |
ey | -0.048 |
city | -0.047 |
classroom | -0.047 |
museum | -0.047 |
Token | you |
Token ID | 345 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
es | +0.030 |
icked | +0.028 |
on | +0.027 |
together | +0.027 |
against | +0.024 |
Neg logit contributions | |
hotel | -0.041 |
museum | -0.040 |
ey | -0.040 |
principal | -0.040 |
city | -0.040 |
Token | help |
Token ID | 1037 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
hotel | +0.029 |
city | +0.028 |
classroom | +0.028 |
ey | +0.028 |
principal | +0.028 |
Neg logit contributions | |
es | -0.043 |
icked | -0.039 |
on | -0.038 |
together | -0.038 |
there | -0.037 |
Token | me |
Token ID | 502 |
Feature activation | +7.89e-03 |
Pos logit contributions | |
es | +0.037 |
icked | +0.036 |
on | +0.033 |
together | +0.033 |
there | +0.032 |
Neg logit contributions | |
hotel | -0.023 |
classroom | -0.022 |
city | -0.022 |
museum | -0.022 |
principal | -0.022 |
Token | repair |
Token ID | 9185 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
hotel | +0.012 |
classroom | +0.012 |
city | +0.012 |
playground | +0.012 |
museum | +0.012 |
Neg logit contributions | |
es | -0.017 |
icked | -0.016 |
together | -0.014 |
on | -0.014 |
there | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
es | +0.029 |
icked | +0.026 |
together | +0.024 |
on | +0.024 |
there | +0.023 |
Neg logit contributions | |
hotel | -0.053 |
ey | -0.051 |
city | -0.051 |
classroom | -0.051 |
principal | -0.051 |
Token | hole |
Token ID | 7604 |
Feature activation | -5.04e-02 |
Pos logit contributions | |
carn | +0.251 |
orchestra | +0.251 |
former | +0.245 |
hotel | +0.243 |
ey | +0.241 |
Neg logit contributions | |
on | -0.387 |
there | -0.372 |
in | -0.357 |
es | -0.344 |
with | -0.333 |
Token | in |
Token ID | 287 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
es | +0.075 |
icked | +0.070 |
ged | +0.053 |
together | +0.053 |
iled | +0.052 |
Neg logit contributions | |
hotel | -0.109 |
classroom | -0.104 |
city | -0.104 |
museum | -0.102 |
playground | -0.102 |
Token | my |
Token ID | 616 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
hotel | +0.049 |
city | +0.047 |
ey | +0.047 |
classroom | +0.047 |
museum | +0.046 |
Neg logit contributions | |
es | -0.022 |
icked | -0.020 |
on | -0.018 |
together | -0.018 |
there | -0.017 |
Token | wall |
Token ID | 3355 |
Feature activation | -2.42e-02 |
Pos logit contributions | |
orchestra | +0.192 |
ey | +0.191 |
Ta | +0.185 |
priest | +0.182 |
carn | +0.179 |
Neg logit contributions | |
on | -0.313 |
there | -0.303 |
es | -0.300 |
in | -0.293 |
together | -0.286 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
es | +0.033 |
icked | +0.031 |
on | +0.026 |
together | +0.026 |
there | +0.025 |
Neg logit contributions | |
hotel | -0.056 |
classroom | -0.055 |
city | -0.054 |
museum | -0.053 |
playground | -0.053 |
Token | sh |
Token ID | 427 |
Feature activation | +1.35e-03 |
Pos logit contributions | |
es | +0.022 |
icked | +0.020 |
together | +0.019 |
on | +0.019 |
there | +0.018 |
Neg logit contributions | |
hotel | -0.015 |
city | -0.014 |
playground | -0.014 |
classroom | -0.014 |
ey | -0.014 |
Token | iver |
Token ID | 1428 |
Feature activation | -7.43e-02 |
Pos logit contributions | |
hotel | +0.002 |
ey | +0.002 |
city | +0.002 |
classroom | +0.002 |
former | +0.002 |
Neg logit contributions | |
es | -0.003 |
there | -0.003 |
on | -0.003 |
icked | -0.003 |
together | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
icked | +0.101 |
es | +0.094 |
there | +0.067 |
together | +0.066 |
able | +0.065 |
Neg logit contributions | |
hotel | -0.178 |
city | -0.169 |
beach | -0.163 |
airport | -0.162 |
playground | -0.161 |
Token | They |
Token ID | 1119 |
Feature activation | -8.38e-03 |
Pos logit contributions | |
es | +0.022 |
icked | +0.020 |
on | +0.019 |
together | +0.018 |
able | +0.017 |
Neg logit contributions | |
hotel | -0.026 |
classroom | -0.026 |
ey | -0.025 |
playground | -0.025 |
city | -0.025 |
Token | wore |
Token ID | 12408 |
Feature activation | -5.09e-02 |
Pos logit contributions | |
es | +0.013 |
together | +0.011 |
icked | +0.011 |
on | +0.011 |
there | +0.010 |
Neg logit contributions | |
hotel | -0.019 |
city | -0.018 |
ey | -0.017 |
classroom | -0.017 |
playground | -0.017 |
Token | hats |
Token ID | 23910 |
Feature activation | -9.76e-02 |
Pos logit contributions | |
es | +0.084 |
icked | +0.077 |
together | +0.069 |
on | +0.068 |
there | +0.066 |
Neg logit contributions | |
hotel | -0.103 |
ey | -0.098 |
city | -0.098 |
playground | -0.097 |
classroom | -0.097 |
Token | and |
Token ID | 290 |
Feature activation | +6.29e-04 |
Pos logit contributions | |
icked | +0.123 |
es | +0.120 |
there | +0.086 |
able | +0.085 |
together | +0.083 |
Neg logit contributions | |
hotel | -0.211 |
city | -0.205 |
playground | -0.203 |
airport | -0.202 |
beach | -0.201 |
Token | coats |
Token ID | 30720 |
Feature activation | -9.61e-02 |
Pos logit contributions | |
hotel | +0.001 |
ey | +0.001 |
principal | +0.001 |
city | +0.001 |
classroom | +0.001 |
Neg logit contributions | |
es | -0.001 |
on | -0.001 |
together | -0.001 |
icked | -0.001 |
there | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
es | +0.113 |
icked | +0.106 |
there | +0.072 |
able | +0.069 |
together | +0.067 |
Neg logit contributions | |
hotel | -0.228 |
playground | -0.223 |
city | -0.221 |
beach | -0.219 |
bus | -0.218 |
Token | but |
Token ID | 475 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
es | +0.031 |
on | +0.029 |
together | +0.028 |
there | +0.028 |
icked | +0.028 |
Neg logit contributions | |
hotel | -0.040 |
ey | -0.039 |
city | -0.038 |
classroom | -0.038 |
principal | -0.038 |
Token | they |
Token ID | 484 |
Feature activation | -4.46e-03 |
Pos logit contributions | |
es | +0.019 |
icked | +0.016 |
together | +0.012 |
able | +0.011 |
on | +0.011 |
Neg logit contributions | |
hotel | -0.042 |
beach | -0.040 |
classroom | -0.040 |
playground | -0.040 |
mall | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
icked | +0.101 |
es | +0.094 |
there | +0.067 |
together | +0.066 |
able | +0.065 |
Neg logit contributions | |
hotel | -0.178 |
city | -0.169 |
beach | -0.163 |
airport | -0.162 |
playground | -0.161 |
Token | They |
Token ID | 1119 |
Feature activation | -8.38e-03 |
Pos logit contributions | |
es | +0.022 |
icked | +0.020 |
on | +0.019 |
together | +0.018 |
able | +0.017 |
Neg logit contributions | |
hotel | -0.026 |
classroom | -0.026 |
ey | -0.025 |
playground | -0.025 |
city | -0.025 |
Token | wore |
Token ID | 12408 |
Feature activation | -5.09e-02 |
Pos logit contributions | |
es | +0.013 |
together | +0.011 |
icked | +0.011 |
on | +0.011 |
there | +0.010 |
Neg logit contributions | |
hotel | -0.019 |
city | -0.018 |
ey | -0.017 |
classroom | -0.017 |
playground | -0.017 |
Token | hats |
Token ID | 23910 |
Feature activation | -9.76e-02 |
Pos logit contributions | |
es | +0.084 |
icked | +0.077 |
together | +0.069 |
on | +0.068 |
there | +0.066 |
Neg logit contributions | |
hotel | -0.103 |
ey | -0.098 |
city | -0.098 |
playground | -0.097 |
classroom | -0.097 |
Token | and |
Token ID | 290 |
Feature activation | +6.29e-04 |
Pos logit contributions | |
icked | +0.123 |
es | +0.120 |
there | +0.086 |
able | +0.085 |
together | +0.083 |
Neg logit contributions | |
hotel | -0.211 |
city | -0.205 |
playground | -0.203 |
airport | -0.202 |
beach | -0.201 |
Token | coats |
Token ID | 30720 |
Feature activation | -9.61e-02 |
Pos logit contributions | |
hotel | +0.001 |
ey | +0.001 |
principal | +0.001 |
city | +0.001 |
classroom | +0.001 |
Neg logit contributions | |
es | -0.001 |
on | -0.001 |
together | -0.001 |
icked | -0.001 |
there | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
es | +0.113 |
icked | +0.106 |
there | +0.072 |
able | +0.069 |
together | +0.067 |
Neg logit contributions | |
hotel | -0.228 |
playground | -0.223 |
city | -0.221 |
beach | -0.219 |
bus | -0.218 |
Token | but |
Token ID | 475 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
es | +0.031 |
on | +0.029 |
together | +0.028 |
there | +0.028 |
icked | +0.028 |
Neg logit contributions | |
hotel | -0.040 |
ey | -0.039 |
city | -0.038 |
classroom | -0.038 |
principal | -0.038 |
Token | they |
Token ID | 484 |
Feature activation | -4.46e-03 |
Pos logit contributions | |
es | +0.019 |
icked | +0.016 |
together | +0.012 |
able | +0.011 |
on | +0.011 |
Neg logit contributions | |
hotel | -0.042 |
beach | -0.040 |
classroom | -0.040 |
playground | -0.040 |
mall | -0.039 |
Token | still |
Token ID | 991 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
es | +0.006 |
together | +0.006 |
icked | +0.005 |
on | +0.005 |
there | +0.005 |
Neg logit contributions | |
hotel | -0.010 |
city | -0.010 |
ey | -0.010 |
classroom | -0.010 |
principal | -0.009 |
Token | felt |
Token ID | 2936 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
es | +0.040 |
icked | +0.036 |
together | +0.036 |
on | +0.034 |
there | +0.033 |
Neg logit contributions | |
hotel | -0.050 |
classroom | -0.047 |
city | -0.047 |
playground | -0.046 |
ey | -0.046 |
Token | looking |
Token ID | 2045 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
es | +0.024 |
icked | +0.023 |
there | +0.020 |
on | +0.020 |
together | +0.020 |
Neg logit contributions | |
hotel | -0.040 |
city | -0.039 |
classroom | -0.039 |
playground | -0.038 |
ey | -0.038 |
Token | at |
Token ID | 379 |
Feature activation | -7.22e-03 |
Pos logit contributions | |
es | +0.013 |
icked | +0.010 |
on | +0.009 |
there | +0.009 |
together | +0.009 |
Neg logit contributions | |
hotel | -0.059 |
ey | -0.057 |
city | -0.057 |
classroom | -0.056 |
principal | -0.056 |
Token | bugs |
Token ID | 11316 |
Feature activation | -5.21e-02 |
Pos logit contributions | |
es | +0.006 |
on | +0.006 |
together | +0.006 |
there | +0.005 |
icked | +0.005 |
Neg logit contributions | |
hotel | -0.020 |
ey | -0.020 |
principal | -0.020 |
classroom | -0.020 |
orchestra | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | -1.38e-03 |
Pos logit contributions | |
icked | +0.055 |
es | +0.048 |
there | +0.039 |
on | +0.038 |
against | +0.037 |
Neg logit contributions | |
hotel | -0.132 |
beach | -0.127 |
playground | -0.127 |
city | -0.127 |
zoo | -0.126 |
Token | wearing |
Token ID | 5762 |
Feature activation | -3.33e-02 |
Pos logit contributions | |
es | +0.002 |
on | +0.002 |
together | +0.002 |
there | +0.002 |
icked | +0.002 |
Neg logit contributions | |
hotel | -0.003 |
principal | -0.003 |
ey | -0.003 |
city | -0.003 |
classroom | -0.003 |
Token | hats |
Token ID | 23910 |
Feature activation | -9.56e-02 |
Pos logit contributions | |
es | +0.049 |
on | +0.044 |
icked | +0.043 |
together | +0.043 |
there | +0.043 |
Neg logit contributions | |
hotel | -0.074 |
ey | -0.072 |
city | -0.071 |
classroom | -0.071 |
principal | -0.071 |
Token | . |
Token ID | 13 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
icked | +0.106 |
es | +0.102 |
iled | +0.074 |
there | +0.072 |
together | +0.065 |
Neg logit contributions | |
hotel | -0.217 |
beach | -0.215 |
airport | -0.211 |
playground | -0.211 |
city | -0.211 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
es | +0.025 |
icked | +0.022 |
on | +0.022 |
there | +0.020 |
together | +0.020 |
Neg logit contributions | |
city | -0.034 |
hotel | -0.034 |
playground | -0.033 |
museum | -0.033 |
zoo | -0.032 |
Token | Once |
Token ID | 7454 |
Feature activation | +3.10e-02 |
Pos logit contributions | |
hotel | +0.024 |
city | +0.023 |
museum | +0.023 |
classroom | +0.022 |
zoo | +0.022 |
Neg logit contributions | |
icked | -0.022 |
es | -0.022 |
able | -0.020 |
together | -0.020 |
instead | -0.020 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
hotel | +0.053 |
classroom | +0.050 |
ey | +0.050 |
city | +0.050 |
museum | +0.048 |
Neg logit contributions | |
es | -0.062 |
icked | -0.057 |
on | -0.056 |
together | -0.055 |
there | -0.052 |
Token | a |
Token ID | 257 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
es | +0.014 |
icked | +0.013 |
together | +0.012 |
on | +0.012 |
there | +0.011 |
Neg logit contributions | |
hotel | -0.026 |
classroom | -0.025 |
city | -0.025 |
ey | -0.025 |
playground | -0.025 |
Token | see |
Token ID | 766 |
Feature activation | -3.52e-02 |
Pos logit contributions | |
es | +0.011 |
on | +0.011 |
there | +0.011 |
together | +0.010 |
in | +0.010 |
Neg logit contributions | |
principal | -0.015 |
ey | -0.015 |
hotel | -0.014 |
former | -0.014 |
orchestra | -0.014 |
Token | some |
Token ID | 617 |
Feature activation | +2.94e-02 |
Pos logit contributions | |
es | +0.045 |
on | +0.041 |
together | +0.040 |
icked | +0.039 |
there | +0.039 |
Neg logit contributions | |
hotel | -0.085 |
ey | -0.083 |
principal | -0.082 |
classroom | -0.082 |
city | -0.082 |
Token | rocks |
Token ID | 12586 |
Feature activation | -7.65e-02 |
Pos logit contributions | |
principal | +0.047 |
hotel | +0.046 |
classroom | +0.045 |
former | +0.044 |
ey | +0.044 |
Neg logit contributions | |
on | -0.060 |
es | -0.060 |
there | -0.058 |
together | -0.056 |
in | -0.055 |
Token | and |
Token ID | 290 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
icked | +0.091 |
es | +0.089 |
together | +0.065 |
able | +0.063 |
against | +0.063 |
Neg logit contributions | |
hotel | -0.173 |
beach | -0.172 |
ey | -0.171 |
playground | -0.170 |
city | -0.170 |
Token | some |
Token ID | 617 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
es | +0.024 |
on | +0.023 |
there | +0.022 |
together | +0.022 |
icked | +0.021 |
Neg logit contributions | |
hotel | -0.021 |
principal | -0.020 |
ey | -0.020 |
classroom | -0.020 |
city | -0.020 |
Token | bats |
Token ID | 19553 |
Feature activation | -9.51e-02 |
Pos logit contributions | |
hotel | +0.041 |
principal | +0.039 |
ey | +0.039 |
classroom | +0.039 |
city | +0.038 |
Neg logit contributions | |
es | -0.055 |
on | -0.053 |
there | -0.051 |
together | -0.051 |
icked | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
icked | +0.104 |
es | +0.104 |
together | +0.077 |
able | +0.073 |
against | +0.070 |
Neg logit contributions | |
hotel | -0.217 |
beach | -0.216 |
playground | -0.215 |
city | -0.213 |
airport | -0.209 |
Token | They |
Token ID | 1119 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
es | +0.032 |
icked | +0.029 |
able | +0.026 |
on | +0.026 |
together | +0.024 |
Neg logit contributions | |
city | -0.036 |
playground | -0.035 |
hotel | -0.035 |
classroom | -0.034 |
museum | -0.034 |
Token | also |
Token ID | 635 |
Feature activation | -3.17e-02 |
Pos logit contributions | |
es | +0.019 |
on | +0.019 |
there | +0.019 |
together | +0.018 |
in | +0.017 |
Neg logit contributions | |
hotel | -0.024 |
principal | -0.024 |
ey | -0.023 |
former | -0.023 |
babys | -0.023 |
Token | see |
Token ID | 766 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
on | +0.046 |
es | +0.045 |
there | +0.045 |
together | +0.042 |
in | +0.041 |
Neg logit contributions | |
hotel | -0.068 |
principal | -0.066 |
former | -0.066 |
ey | -0.066 |
babys | -0.065 |
Token | a |
Token ID | 257 |
Feature activation | +5.03e-02 |
Pos logit contributions | |
es | +0.051 |
on | +0.046 |
together | +0.045 |
there | +0.045 |
icked | +0.044 |
Neg logit contributions | |
hotel | -0.087 |
ey | -0.086 |
principal | -0.085 |
classroom | -0.084 |
city | -0.084 |
Token | She |
Token ID | 1375 |
Feature activation | -3.97e-03 |
Pos logit contributions | |
es | +0.021 |
on | +0.018 |
together | +0.018 |
icked | +0.018 |
there | +0.018 |
Neg logit contributions | |
hotel | -0.035 |
ey | -0.034 |
classroom | -0.034 |
city | -0.034 |
principal | -0.034 |
Token | put |
Token ID | 1234 |
Feature activation | -2.97e-02 |
Pos logit contributions | |
es | +0.006 |
on | +0.005 |
icked | +0.005 |
together | +0.005 |
there | +0.005 |
Neg logit contributions | |
hotel | -0.009 |
ey | -0.009 |
city | -0.009 |
classroom | -0.009 |
principal | -0.009 |
Token | on |
Token ID | 319 |
Feature activation | -3.85e-02 |
Pos logit contributions | |
es | +0.020 |
icked | +0.014 |
together | +0.012 |
instead | +0.009 |
there | +0.009 |
Neg logit contributions | |
hotel | -0.086 |
ey | -0.084 |
mirror | -0.083 |
city | -0.083 |
classroom | -0.082 |
Token | her |
Token ID | 607 |
Feature activation | +7.68e-02 |
Pos logit contributions | |
es | +0.038 |
on | +0.037 |
there | +0.034 |
together | +0.034 |
icked | +0.030 |
Neg logit contributions | |
hotel | -0.103 |
ey | -0.102 |
principal | -0.102 |
city | -0.099 |
classroom | -0.099 |
Token | rain |
Token ID | 6290 |
Feature activation | -3.92e-02 |
Pos logit contributions | |
hotel | +0.121 |
principal | +0.121 |
ey | +0.120 |
city | +0.117 |
museum | +0.116 |
Neg logit contributions | |
on | -0.157 |
es | -0.153 |
there | -0.151 |
together | -0.147 |
icked | -0.142 |
Token | coat |
Token ID | 31434 |
Feature activation | -9.50e-02 |
Pos logit contributions | |
es | +0.064 |
icked | +0.059 |
on | +0.057 |
together | +0.056 |
there | +0.056 |
Neg logit contributions | |
hotel | -0.081 |
classroom | -0.078 |
ey | -0.077 |
city | -0.077 |
playground | -0.076 |
Token | and |
Token ID | 290 |
Feature activation | -2.17e-02 |
Pos logit contributions | |
es | +0.115 |
icked | +0.109 |
together | +0.084 |
there | +0.083 |
on | +0.076 |
Neg logit contributions | |
hotel | -0.233 |
playground | -0.229 |
classroom | -0.227 |
city | -0.225 |
mall | -0.222 |
Token | boots |
Token ID | 14412 |
Feature activation | -8.67e-02 |
Pos logit contributions | |
es | +0.037 |
icked | +0.034 |
together | +0.033 |
on | +0.033 |
there | +0.032 |
Neg logit contributions | |
hotel | -0.043 |
ey | -0.041 |
classroom | -0.041 |
city | -0.041 |
principal | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
es | +0.098 |
icked | +0.088 |
there | +0.070 |
together | +0.069 |
on | +0.065 |
Neg logit contributions | |
hotel | -0.209 |
playground | -0.205 |
beach | -0.201 |
salon | -0.200 |
city | -0.200 |
Token | She |
Token ID | 1375 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
es | +0.024 |
icked | +0.021 |
on | +0.020 |
together | +0.020 |
there | +0.020 |
Neg logit contributions | |
hotel | -0.033 |
classroom | -0.032 |
ey | -0.032 |
principal | -0.032 |
playground | -0.032 |
Token | did |
Token ID | 750 |
Feature activation | -5.26e-02 |
Pos logit contributions | |
es | +0.015 |
on | +0.014 |
together | +0.013 |
icked | +0.013 |
there | +0.013 |
Neg logit contributions | |
hotel | -0.023 |
ey | -0.023 |
city | -0.023 |
principal | -0.023 |
classroom | -0.023 |
Token | anymore |
Token ID | 7471 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
es | +0.025 |
icked | +0.023 |
iled | +0.019 |
hot | +0.017 |
dd | +0.017 |
Neg logit contributions | |
hotel | -0.036 |
beach | -0.035 |
playground | -0.033 |
museum | -0.033 |
classroom | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -2.66e-02 |
Pos logit contributions | |
icked | +0.039 |
es | +0.033 |
pped | +0.030 |
bled | +0.029 |
iled | +0.027 |
Neg logit contributions | |
classroom | -0.052 |
hotel | -0.051 |
playground | -0.050 |
museum | -0.050 |
beach | -0.050 |
Token | He |
Token ID | 679 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
es | +0.040 |
on | +0.035 |
icked | +0.035 |
together | +0.034 |
there | +0.034 |
Neg logit contributions | |
hotel | -0.059 |
ey | -0.058 |
classroom | -0.058 |
city | -0.058 |
principal | -0.057 |
Token | dug |
Token ID | 18735 |
Feature activation | -6.59e-02 |
Pos logit contributions | |
es | +0.027 |
on | +0.024 |
icked | +0.023 |
together | +0.023 |
there | +0.023 |
Neg logit contributions | |
hotel | -0.046 |
ey | -0.044 |
city | -0.044 |
classroom | -0.044 |
principal | -0.044 |
Token | a |
Token ID | 257 |
Feature activation | +4.73e-02 |
Pos logit contributions | |
es | +0.082 |
icked | +0.073 |
together | +0.059 |
on | +0.057 |
there | +0.056 |
Neg logit contributions | |
hotel | -0.157 |
ey | -0.154 |
city | -0.151 |
museum | -0.150 |
playground | -0.149 |
Token | hole |
Token ID | 7604 |
Feature activation | -9.48e-02 |
Pos logit contributions | |
principal | +0.099 |
orchestra | +0.093 |
pastor | +0.093 |
priest | +0.092 |
opl | +0.092 |
Neg logit contributions | |
on | -0.082 |
there | -0.079 |
together | -0.078 |
in | -0.076 |
with | -0.075 |
Token | and |
Token ID | 290 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
es | +0.117 |
icked | +0.099 |
rums | +0.071 |
mar | +0.065 |
hot | +0.063 |
Neg logit contributions | |
doorway | -0.217 |
hotel | -0.213 |
playground | -0.206 |
ey | -0.206 |
beach | -0.206 |
Token | carefully |
Token ID | 7773 |
Feature activation | -4.73e-02 |
Pos logit contributions | |
es | +0.045 |
icked | +0.040 |
on | +0.038 |
together | +0.036 |
there | +0.035 |
Neg logit contributions | |
hotel | -0.057 |
beach | -0.055 |
playground | -0.055 |
museum | -0.053 |
doorway | -0.053 |
Token | placed |
Token ID | 4624 |
Feature activation | -4.04e-02 |
Pos logit contributions | |
es | +0.087 |
icked | +0.072 |
together | +0.069 |
on | +0.068 |
against | +0.065 |
Neg logit contributions | |
beach | -0.081 |
hotel | -0.079 |
playground | -0.078 |
doorway | -0.077 |
ey | -0.074 |
Token | the |
Token ID | 262 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
es | +0.036 |
icked | +0.028 |
together | +0.024 |
on | +0.021 |
there | +0.019 |
Neg logit contributions | |
hotel | -0.112 |
ey | -0.110 |
city | -0.109 |
museum | -0.109 |
playground | -0.106 |
Token | skull |
Token ID | 14511 |
Feature activation | -4.88e-02 |
Pos logit contributions | |
hotel | +0.229 |
principal | +0.217 |
salon | +0.213 |
former | +0.211 |
orchestra | +0.210 |
Neg logit contributions | |
es | -0.332 |
on | -0.329 |
there | -0.323 |
together | -0.305 |
icked | -0.302 |
Token | She |
Token ID | 1375 |
Feature activation | -6.92e-03 |
Pos logit contributions | |
es | +0.011 |
icked | +0.010 |
on | +0.010 |
together | +0.009 |
there | +0.009 |
Neg logit contributions | |
hotel | -0.014 |
city | -0.014 |
ey | -0.014 |
playground | -0.014 |
classroom | -0.014 |
Token | put |
Token ID | 1234 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
es | +0.009 |
on | +0.008 |
together | +0.008 |
there | +0.008 |
icked | +0.008 |
Neg logit contributions | |
hotel | -0.016 |
ey | -0.016 |
principal | -0.016 |
city | -0.016 |
classroom | -0.016 |
Token | on |
Token ID | 319 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
es | +0.027 |
icked | +0.022 |
together | +0.019 |
instead | +0.017 |
there | +0.017 |
Neg logit contributions | |
hotel | -0.074 |
ey | -0.074 |
classroom | -0.072 |
mirror | -0.072 |
city | -0.071 |
Token | her |
Token ID | 607 |
Feature activation | +8.24e-02 |
Pos logit contributions | |
es | +0.037 |
on | +0.034 |
together | +0.033 |
there | +0.032 |
icked | +0.031 |
Neg logit contributions | |
hotel | -0.086 |
principal | -0.084 |
ey | -0.084 |
city | -0.083 |
classroom | -0.083 |
Token | rain |
Token ID | 6290 |
Feature activation | -3.86e-02 |
Pos logit contributions | |
principal | +0.140 |
ey | +0.137 |
hotel | +0.137 |
museum | +0.134 |
priest | +0.132 |
Neg logit contributions | |
on | -0.165 |
there | -0.158 |
es | -0.151 |
together | -0.151 |
in | -0.150 |
Token | coat |
Token ID | 31434 |
Feature activation | -9.48e-02 |
Pos logit contributions | |
es | +0.082 |
icked | +0.077 |
on | +0.074 |
there | +0.074 |
together | +0.073 |
Neg logit contributions | |
hotel | -0.061 |
classroom | -0.058 |
city | -0.058 |
playground | -0.057 |
ey | -0.057 |
Token | and |
Token ID | 290 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
es | +0.110 |
icked | +0.105 |
there | +0.080 |
together | +0.079 |
on | +0.073 |
Neg logit contributions | |
hotel | -0.235 |
playground | -0.231 |
city | -0.228 |
classroom | -0.227 |
mall | -0.225 |
Token | ran |
Token ID | 4966 |
Feature activation | -4.70e-02 |
Pos logit contributions | |
es | +0.035 |
on | +0.032 |
icked | +0.032 |
together | +0.032 |
there | +0.031 |
Neg logit contributions | |
hotel | -0.037 |
ey | -0.036 |
city | -0.035 |
classroom | -0.035 |
principal | -0.035 |
Token | to |
Token ID | 284 |
Feature activation | -7.47e-03 |
Pos logit contributions | |
es | +0.057 |
icked | +0.050 |
on | +0.047 |
together | +0.047 |
there | +0.046 |
Neg logit contributions | |
hotel | -0.118 |
ey | -0.114 |
classroom | -0.114 |
city | -0.113 |
playground | -0.112 |
Token | the |
Token ID | 262 |
Feature activation | +9.90e-02 |
Pos logit contributions | |
on | +0.007 |
es | +0.007 |
there | +0.007 |
together | +0.007 |
icked | +0.006 |
Neg logit contributions | |
hotel | -0.020 |
ey | -0.020 |
principal | -0.020 |
city | -0.019 |
priest | -0.019 |
Token | window |
Token ID | 4324 |
Feature activation | -6.66e-02 |
Pos logit contributions | |
igsaw | +0.163 |
opl | +0.158 |
ey | +0.158 |
actor | +0.157 |
Ta | +0.155 |
Neg logit contributions | |
on | -0.206 |
there | -0.203 |
in | -0.200 |
together | -0.183 |
with | -0.183 |
Token | They |
Token ID | 1119 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
es | +0.022 |
on | +0.019 |
together | +0.019 |
icked | +0.019 |
there | +0.018 |
Neg logit contributions | |
hotel | -0.028 |
ey | -0.027 |
principal | -0.027 |
classroom | -0.027 |
city | -0.027 |
Token | wear |
Token ID | 5806 |
Feature activation | -4.91e-02 |
Pos logit contributions | |
on | +0.023 |
es | +0.023 |
there | +0.023 |
together | +0.021 |
in | +0.021 |
Neg logit contributions | |
ey | -0.028 |
principal | -0.028 |
former | -0.028 |
igsaw | -0.027 |
classroom | -0.027 |
Token | big |
Token ID | 1263 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
es | +0.077 |
on | +0.070 |
icked | +0.069 |
together | +0.069 |
there | +0.068 |
Neg logit contributions | |
hotel | -0.104 |
ey | -0.101 |
city | -0.099 |
classroom | -0.099 |
principal | -0.099 |
Token | hats |
Token ID | 23910 |
Feature activation | -9.34e-02 |
Pos logit contributions | |
hotel | +0.021 |
principal | +0.020 |
former | +0.020 |
ey | +0.020 |
classroom | +0.020 |
Neg logit contributions | |
es | -0.029 |
on | -0.028 |
there | -0.027 |
together | -0.027 |
icked | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
icked | +0.125 |
es | +0.123 |
able | +0.088 |
rums | +0.087 |
there | +0.086 |
Neg logit contributions | |
hotel | -0.192 |
airport | -0.189 |
bus | -0.189 |
city | -0.188 |
playground | -0.188 |
Token | coats |
Token ID | 30720 |
Feature activation | -9.45e-02 |
Pos logit contributions | |
es | +0.026 |
on | +0.025 |
together | +0.024 |
there | +0.024 |
icked | +0.024 |
Neg logit contributions | |
hotel | -0.023 |
ey | -0.023 |
principal | -0.022 |
city | -0.022 |
classroom | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
es | +0.115 |
icked | +0.107 |
there | +0.076 |
able | +0.076 |
together | +0.071 |
Neg logit contributions | |
hotel | -0.220 |
bus | -0.217 |
playground | -0.216 |
city | -0.215 |
classroom | -0.214 |
Token | They |
Token ID | 1119 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
es | +0.019 |
icked | +0.016 |
on | +0.016 |
together | +0.016 |
there | +0.015 |
Neg logit contributions | |
hotel | -0.023 |
classroom | -0.022 |
city | -0.022 |
principal | -0.022 |
ey | -0.021 |
Token | have |
Token ID | 423 |
Feature activation | -9.70e-03 |
Pos logit contributions | |
on | +0.029 |
es | +0.027 |
there | +0.027 |
together | +0.026 |
in | +0.026 |
Neg logit contributions | |
ey | -0.033 |
former | -0.032 |
principal | -0.032 |
doorway | -0.032 |
igsaw | -0.032 |
Token | a |
Token ID | 257 |
Feature activation | +3.62e-02 |
Pos logit contributions | |
es | +0.017 |
icked | +0.016 |
on | +0.016 |
together | +0.015 |
there | +0.015 |
Neg logit contributions | |
hotel | -0.019 |
ey | -0.018 |
city | -0.018 |
classroom | -0.018 |
principal | -0.018 |
Token | dog |
Token ID | 3290 |
Feature activation | -5.24e-02 |
Pos logit contributions | |
ey | +0.065 |
opl | +0.060 |
former | +0.060 |
principal | +0.059 |
Ta | +0.058 |
Neg logit contributions | |
on | -0.077 |
there | -0.071 |
together | -0.069 |
in | -0.063 |
, | -0.063 |
Token | mer |
Token ID | 647 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
es | +0.005 |
icked | +0.005 |
on | +0.004 |
together | +0.004 |
there | +0.004 |
Neg logit contributions | |
hotel | -0.003 |
city | -0.003 |
museum | -0.003 |
ostr | -0.003 |
ey | -0.003 |
Token | Joe |
Token ID | 5689 |
Feature activation | -2.24e-02 |
Pos logit contributions | |
hotel | +0.023 |
city | +0.023 |
playground | +0.022 |
principal | +0.022 |
museum | +0.022 |
Neg logit contributions | |
es | -0.015 |
icked | -0.013 |
together | -0.013 |
on | -0.013 |
there | -0.012 |
Token | dug |
Token ID | 18735 |
Feature activation | -6.01e-02 |
Pos logit contributions | |
es | +0.034 |
icked | +0.033 |
on | +0.031 |
together | +0.031 |
there | +0.029 |
Neg logit contributions | |
hotel | -0.046 |
city | -0.045 |
playground | -0.044 |
museum | -0.044 |
beach | -0.043 |
Token | a |
Token ID | 257 |
Feature activation | +5.24e-02 |
Pos logit contributions | |
es | +0.076 |
icked | +0.068 |
together | +0.060 |
on | +0.056 |
there | +0.056 |
Neg logit contributions | |
hotel | -0.144 |
ey | -0.138 |
city | -0.138 |
museum | -0.135 |
playground | -0.135 |
Token | big |
Token ID | 1263 |
Feature activation | +4.43e-02 |
Pos logit contributions | |
principal | +0.105 |
opl | +0.100 |
orchestra | +0.096 |
actor | +0.093 |
Phantom | +0.092 |
Neg logit contributions | |
on | -0.097 |
there | -0.095 |
together | -0.092 |
in | -0.089 |
with | -0.087 |
Token | hole |
Token ID | 7604 |
Feature activation | -9.44e-02 |
Pos logit contributions | |
principal | +0.070 |
hotel | +0.070 |
ey | +0.068 |
classroom | +0.067 |
city | +0.066 |
Neg logit contributions | |
es | -0.090 |
on | -0.088 |
there | -0.086 |
together | -0.085 |
icked | -0.081 |
Token | with |
Token ID | 351 |
Feature activation | -2.72e-03 |
Pos logit contributions | |
es | +0.106 |
icked | +0.090 |
rums | +0.067 |
hot | +0.058 |
together | +0.057 |
Neg logit contributions | |
doorway | -0.227 |
hotel | -0.223 |
beach | -0.218 |
playground | -0.216 |
hallway | -0.215 |
Token | a |
Token ID | 257 |
Feature activation | +3.28e-02 |
Pos logit contributions | |
es | +0.003 |
on | +0.003 |
there | +0.003 |
together | +0.003 |
icked | +0.003 |
Neg logit contributions | |
hotel | -0.007 |
principal | -0.006 |
ey | -0.006 |
classroom | -0.006 |
city | -0.006 |
Token | shovel |
Token ID | 33677 |
Feature activation | -8.19e-02 |
Pos logit contributions | |
principal | +0.056 |
hotel | +0.055 |
orchestra | +0.053 |
ey | +0.052 |
classroom | +0.052 |
Neg logit contributions | |
on | -0.065 |
there | -0.063 |
es | -0.062 |
together | -0.060 |
in | -0.057 |
Token | and |
Token ID | 290 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
es | +0.094 |
icked | +0.081 |
together | +0.078 |
on | +0.063 |
there | +0.063 |
Neg logit contributions | |
hotel | -0.193 |
ey | -0.186 |
beach | -0.186 |
city | -0.183 |
playground | -0.183 |
Token | put |
Token ID | 1234 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
es | +0.035 |
icked | +0.031 |
on | +0.030 |
together | +0.030 |
there | +0.029 |
Neg logit contributions | |
hotel | -0.057 |
ey | -0.055 |
playground | -0.055 |
city | -0.055 |
classroom | -0.054 |
Token | put |
Token ID | 1234 |
Feature activation | -2.89e-02 |
Pos logit contributions | |
es | +0.016 |
on | +0.014 |
together | +0.014 |
icked | +0.014 |
there | +0.014 |
Neg logit contributions | |
hotel | -0.026 |
ey | -0.025 |
city | -0.025 |
classroom | -0.025 |
principal | -0.025 |
Token | on |
Token ID | 319 |
Feature activation | -3.30e-02 |
Pos logit contributions | |
es | +0.022 |
icked | +0.015 |
together | +0.012 |
iled | +0.010 |
instead | +0.010 |
Neg logit contributions | |
mirror | -0.081 |
ey | -0.080 |
hotel | -0.079 |
classroom | -0.078 |
mall | -0.078 |
Token | her |
Token ID | 607 |
Feature activation | +7.60e-02 |
Pos logit contributions | |
es | +0.033 |
on | +0.032 |
together | +0.029 |
there | +0.029 |
icked | +0.026 |
Neg logit contributions | |
hotel | -0.088 |
ey | -0.087 |
principal | -0.086 |
city | -0.085 |
orchestra | -0.085 |
Token | yellow |
Token ID | 7872 |
Feature activation | +8.07e-02 |
Pos logit contributions | |
hotel | +0.119 |
ey | +0.116 |
principal | +0.116 |
city | +0.113 |
museum | +0.111 |
Neg logit contributions | |
es | -0.157 |
on | -0.155 |
there | -0.150 |
together | -0.148 |
icked | -0.146 |
Token | rain |
Token ID | 6290 |
Feature activation | -4.20e-02 |
Pos logit contributions | |
hotel | +0.115 |
principal | +0.110 |
ey | +0.110 |
city | +0.108 |
classroom | +0.106 |
Neg logit contributions | |
es | -0.180 |
on | -0.174 |
together | -0.169 |
there | -0.168 |
icked | -0.167 |
Token | coat |
Token ID | 31434 |
Feature activation | -9.43e-02 |
Pos logit contributions | |
es | +0.090 |
icked | +0.084 |
there | +0.081 |
on | +0.081 |
together | +0.081 |
Neg logit contributions | |
hotel | -0.065 |
classroom | -0.062 |
city | -0.062 |
ey | -0.061 |
playground | -0.061 |
Token | and |
Token ID | 290 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
es | +0.117 |
icked | +0.110 |
there | +0.087 |
together | +0.086 |
on | +0.074 |
Neg logit contributions | |
hotel | -0.225 |
playground | -0.222 |
classroom | -0.220 |
city | -0.219 |
airport | -0.219 |
Token | boots |
Token ID | 14412 |
Feature activation | -8.44e-02 |
Pos logit contributions | |
es | +0.052 |
icked | +0.047 |
together | +0.045 |
on | +0.045 |
there | +0.044 |
Neg logit contributions | |
hotel | -0.063 |
classroom | -0.060 |
city | -0.060 |
ey | -0.060 |
playground | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
es | +0.092 |
icked | +0.077 |
there | +0.067 |
together | +0.064 |
on | +0.058 |
Neg logit contributions | |
hotel | -0.206 |
babys | -0.202 |
playground | -0.201 |
salon | -0.200 |
bus | -0.200 |
Token | went |
Token ID | 1816 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
es | +0.042 |
icked | +0.037 |
together | +0.036 |
on | +0.036 |
there | +0.035 |
Neg logit contributions | |
hotel | -0.058 |
city | -0.056 |
classroom | -0.056 |
ey | -0.055 |
playground | -0.055 |
Token | outside |
Token ID | 2354 |
Feature activation | -6.20e-02 |
Pos logit contributions | |
es | +0.041 |
icked | +0.035 |
together | +0.032 |
there | +0.032 |
on | +0.031 |
Neg logit contributions | |
hotel | -0.084 |
classroom | -0.082 |
city | -0.080 |
ey | -0.080 |
playground | -0.080 |
Token | She |
Token ID | 1375 |
Feature activation | -9.16e-03 |
Pos logit contributions | |
es | +0.033 |
together | +0.029 |
on | +0.029 |
icked | +0.029 |
there | +0.028 |
Neg logit contributions | |
hotel | -0.056 |
ey | -0.054 |
classroom | -0.054 |
city | -0.054 |
principal | -0.053 |
Token | puts |
Token ID | 7584 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
es | +0.016 |
on | +0.015 |
there | +0.015 |
icked | +0.014 |
in | +0.014 |
Neg logit contributions | |
hotel | -0.018 |
principal | -0.017 |
former | -0.017 |
igsaw | -0.017 |
priest | -0.017 |
Token | on |
Token ID | 319 |
Feature activation | -4.04e-02 |
Pos logit contributions | |
es | +0.021 |
icked | +0.014 |
together | +0.011 |
there | +0.008 |
instead | +0.008 |
Neg logit contributions | |
hotel | -0.102 |
ey | -0.100 |
classroom | -0.099 |
city | -0.098 |
mirror | -0.098 |
Token | her |
Token ID | 607 |
Feature activation | +7.18e-02 |
Pos logit contributions | |
es | +0.043 |
on | +0.040 |
together | +0.038 |
there | +0.038 |
icked | +0.036 |
Neg logit contributions | |
hotel | -0.106 |
ey | -0.104 |
principal | -0.103 |
city | -0.102 |
classroom | -0.101 |
Token | rain |
Token ID | 6290 |
Feature activation | -3.92e-02 |
Pos logit contributions | |
hotel | +0.116 |
principal | +0.115 |
ey | +0.115 |
former | +0.113 |
orchestra | +0.112 |
Neg logit contributions | |
on | -0.145 |
there | -0.140 |
es | -0.133 |
in | -0.131 |
together | -0.130 |
Token | coat |
Token ID | 31434 |
Feature activation | -9.42e-02 |
Pos logit contributions | |
es | +0.081 |
icked | +0.076 |
together | +0.074 |
on | +0.074 |
there | +0.073 |
Neg logit contributions | |
hotel | -0.064 |
classroom | -0.061 |
ey | -0.060 |
city | -0.060 |
principal | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
es | +0.103 |
icked | +0.099 |
together | +0.080 |
there | +0.080 |
on | +0.074 |
Neg logit contributions | |
hotel | -0.238 |
classroom | -0.231 |
playground | -0.231 |
city | -0.231 |
principal | -0.230 |
Token | boots |
Token ID | 14412 |
Feature activation | -8.49e-02 |
Pos logit contributions | |
es | +0.048 |
on | +0.045 |
icked | +0.044 |
together | +0.044 |
there | +0.043 |
Neg logit contributions | |
hotel | -0.042 |
ey | -0.040 |
city | -0.040 |
classroom | -0.040 |
principal | -0.039 |
Token | and |
Token ID | 290 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
es | +0.092 |
icked | +0.084 |
there | +0.070 |
together | +0.067 |
on | +0.064 |
Neg logit contributions | |
hotel | -0.207 |
playground | -0.200 |
salon | -0.199 |
city | -0.199 |
principal | -0.197 |
Token | goes |
Token ID | 2925 |
Feature activation | -3.33e-02 |
Pos logit contributions | |
es | +0.044 |
on | +0.040 |
icked | +0.040 |
together | +0.040 |
there | +0.039 |
Neg logit contributions | |
hotel | -0.046 |
ey | -0.045 |
city | -0.044 |
classroom | -0.044 |
principal | -0.044 |
Token | to |
Token ID | 284 |
Feature activation | -7.66e-03 |
Pos logit contributions | |
es | +0.038 |
icked | +0.033 |
on | +0.032 |
together | +0.031 |
there | +0.030 |
Neg logit contributions | |
hotel | -0.086 |
ey | -0.083 |
classroom | -0.083 |
city | -0.083 |
principal | -0.082 |
Token | on |
Token ID | 319 |
Feature activation | -3.07e-02 |
Pos logit contributions | |
es | +0.016 |
icked | +0.011 |
together | +0.008 |
instead | +0.006 |
iled | +0.006 |
Neg logit contributions | |
hotel | -0.075 |
ey | -0.073 |
classroom | -0.072 |
museum | -0.072 |
city | -0.071 |
Token | his |
Token ID | 465 |
Feature activation | +8.36e-02 |
Pos logit contributions | |
es | +0.032 |
on | +0.032 |
there | +0.030 |
together | +0.029 |
in | +0.026 |
Neg logit contributions | |
hotel | -0.079 |
ey | -0.078 |
principal | -0.078 |
priest | -0.077 |
city | -0.077 |
Token | different |
Token ID | 1180 |
Feature activation | +9.65e-02 |
Pos logit contributions | |
ey | +0.130 |
hotel | +0.129 |
principal | +0.128 |
priest | +0.126 |
former | +0.126 |
Neg logit contributions | |
on | -0.173 |
there | -0.165 |
es | -0.162 |
together | -0.159 |
in | -0.157 |
Token | colored |
Token ID | 16396 |
Feature activation | +5.79e-02 |
Pos logit contributions | |
principal | +0.156 |
hotel | +0.155 |
ey | +0.153 |
priest | +0.151 |
former | +0.149 |
Neg logit contributions | |
on | -0.190 |
es | -0.183 |
there | -0.181 |
together | -0.178 |
in | -0.173 |
Token | swim |
Token ID | 9422 |
Feature activation | -4.34e-02 |
Pos logit contributions | |
hotel | +0.085 |
principal | +0.082 |
former | +0.081 |
ey | +0.081 |
city | +0.080 |
Neg logit contributions | |
es | -0.126 |
on | -0.125 |
there | -0.119 |
together | -0.118 |
icked | -0.114 |
Token | suit |
Token ID | 6063 |
Feature activation | -9.41e-02 |
Pos logit contributions | |
es | +0.090 |
icked | +0.084 |
on | +0.081 |
together | +0.081 |
there | +0.080 |
Neg logit contributions | |
hotel | -0.071 |
classroom | -0.067 |
city | -0.066 |
ey | -0.066 |
playground | -0.066 |
Token | and |
Token ID | 290 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
es | +0.114 |
icked | +0.106 |
together | +0.088 |
there | +0.080 |
on | +0.075 |
Neg logit contributions | |
hotel | -0.222 |
beach | -0.221 |
playground | -0.220 |
classroom | -0.212 |
airport | -0.211 |
Token | ran |
Token ID | 4966 |
Feature activation | -4.51e-02 |
Pos logit contributions | |
es | +0.030 |
on | +0.029 |
there | +0.027 |
together | +0.027 |
icked | +0.027 |
Neg logit contributions | |
hotel | -0.047 |
ey | -0.046 |
principal | -0.045 |
city | -0.045 |
classroom | -0.045 |
Token | out |
Token ID | 503 |
Feature activation | -5.06e-02 |
Pos logit contributions | |
es | +0.036 |
on | +0.033 |
together | +0.031 |
there | +0.030 |
icked | +0.029 |
Neg logit contributions | |
hotel | -0.129 |
ey | -0.128 |
principal | -0.126 |
city | -0.126 |
classroom | -0.125 |
Token | to |
Token ID | 284 |
Feature activation | -2.73e-03 |
Pos logit contributions | |
es | +0.054 |
icked | +0.048 |
together | +0.040 |
on | +0.040 |
there | +0.040 |
Neg logit contributions | |
hotel | -0.133 |
ey | -0.129 |
classroom | -0.129 |
city | -0.128 |
playground | -0.126 |
Token | the |
Token ID | 262 |
Feature activation | +9.86e-02 |
Pos logit contributions | |
on | +0.003 |
there | +0.003 |
together | +0.003 |
es | +0.003 |
in | +0.003 |
Neg logit contributions | |
ey | -0.007 |
principal | -0.007 |
former | -0.007 |
hotel | -0.007 |
priest | -0.007 |
Token | on |
Token ID | 319 |
Feature activation | -3.50e-02 |
Pos logit contributions | |
es | +0.028 |
icked | +0.021 |
able | +0.019 |
instead | +0.016 |
iled | +0.015 |
Neg logit contributions | |
hotel | -0.088 |
mall | -0.083 |
city | -0.082 |
beach | -0.082 |
mirror | -0.081 |
Token | their |
Token ID | 511 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
es | +0.050 |
on | +0.047 |
together | +0.046 |
there | +0.044 |
icked | +0.043 |
Neg logit contributions | |
hotel | -0.079 |
ey | -0.078 |
principal | -0.078 |
classroom | -0.076 |
city | -0.076 |
Token | swim |
Token ID | 9422 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
hotel | +0.153 |
ey | +0.149 |
principal | +0.146 |
city | +0.145 |
classroom | +0.144 |
Neg logit contributions | |
es | -0.215 |
on | -0.205 |
together | -0.201 |
there | -0.200 |
icked | -0.199 |
Token | suits |
Token ID | 14803 |
Feature activation | -7.48e-02 |
Pos logit contributions | |
es | +0.062 |
icked | +0.058 |
on | +0.057 |
together | +0.057 |
there | +0.056 |
Neg logit contributions | |
hotel | -0.045 |
ey | -0.042 |
classroom | -0.042 |
city | -0.042 |
museum | -0.041 |
Token | and |
Token ID | 290 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
es | +0.094 |
icked | +0.094 |
there | +0.065 |
together | +0.065 |
able | +0.065 |
Neg logit contributions | |
hotel | -0.174 |
beach | -0.170 |
city | -0.166 |
playground | -0.166 |
airport | -0.165 |
Token | hats |
Token ID | 23910 |
Feature activation | -9.40e-02 |
Pos logit contributions | |
es | +0.034 |
on | +0.032 |
icked | +0.031 |
together | +0.031 |
there | +0.031 |
Neg logit contributions | |
hotel | -0.030 |
ey | -0.029 |
city | -0.028 |
classroom | -0.028 |
principal | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -6.39e-03 |
Pos logit contributions | |
es | +0.110 |
icked | +0.106 |
iled | +0.076 |
there | +0.074 |
together | +0.073 |
Neg logit contributions | |
hotel | -0.215 |
beach | -0.210 |
bus | -0.207 |
airport | -0.205 |
playground | -0.205 |
Token | They |
Token ID | 1119 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
es | +0.012 |
icked | +0.011 |
on | +0.010 |
together | +0.010 |
able | +0.010 |
Neg logit contributions | |
hotel | -0.012 |
city | -0.011 |
playground | -0.011 |
mall | -0.011 |
classroom | -0.011 |
Token | get |
Token ID | 651 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
es | +0.018 |
on | +0.018 |
there | +0.017 |
together | +0.016 |
icked | +0.016 |
Neg logit contributions | |
ey | -0.020 |
principal | -0.020 |
classroom | -0.019 |
former | -0.019 |
hotel | -0.019 |
Token | in |
Token ID | 287 |
Feature activation | -9.91e-03 |
Pos logit contributions | |
es | +0.033 |
icked | +0.028 |
together | +0.023 |
instead | +0.023 |
on | +0.023 |
Neg logit contributions | |
hotel | -0.057 |
city | -0.052 |
beach | -0.052 |
classroom | -0.052 |
babys | -0.052 |
Token | the |
Token ID | 262 |
Feature activation | +9.58e-02 |
Pos logit contributions | |
es | +0.011 |
icked | +0.010 |
on | +0.010 |
together | +0.009 |
there | +0.009 |
Neg logit contributions | |
hotel | -0.026 |
ey | -0.025 |
city | -0.025 |
classroom | -0.025 |
principal | -0.024 |
Token | his |
Token ID | 465 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
es | +0.036 |
on | +0.035 |
there | +0.033 |
together | +0.033 |
icked | +0.030 |
Neg logit contributions | |
hotel | -0.096 |
ey | -0.093 |
principal | -0.092 |
orchestra | -0.092 |
priest | -0.092 |
Token | rain |
Token ID | 6290 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
hotel | +0.144 |
city | +0.135 |
ey | +0.132 |
priest | +0.132 |
principal | +0.130 |
Neg logit contributions | |
on | -0.227 |
es | -0.226 |
there | -0.218 |
together | -0.210 |
in | -0.209 |
Token | boots |
Token ID | 14412 |
Feature activation | -7.61e-02 |
Pos logit contributions | |
es | +0.059 |
icked | +0.055 |
on | +0.054 |
together | +0.054 |
there | +0.053 |
Neg logit contributions | |
hotel | -0.046 |
classroom | -0.045 |
ey | -0.044 |
city | -0.044 |
playground | -0.044 |
Token | and |
Token ID | 290 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
es | +0.087 |
icked | +0.081 |
there | +0.067 |
together | +0.063 |
on | +0.062 |
Neg logit contributions | |
hotel | -0.183 |
city | -0.182 |
playground | -0.181 |
classroom | -0.181 |
principal | -0.178 |
Token | rain |
Token ID | 6290 |
Feature activation | -5.47e-02 |
Pos logit contributions | |
es | +0.050 |
icked | +0.044 |
on | +0.043 |
together | +0.042 |
there | +0.042 |
Neg logit contributions | |
hotel | -0.073 |
classroom | -0.071 |
city | -0.071 |
ey | -0.070 |
playground | -0.069 |
Token | coat |
Token ID | 31434 |
Feature activation | -9.39e-02 |
Pos logit contributions | |
es | +0.097 |
icked | +0.090 |
there | +0.086 |
on | +0.084 |
together | +0.083 |
Neg logit contributions | |
hotel | -0.102 |
playground | -0.099 |
classroom | -0.098 |
city | -0.097 |
hallway | -0.095 |
Token | . |
Token ID | 13 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
es | +0.097 |
icked | +0.089 |
there | +0.070 |
together | +0.064 |
on | +0.057 |
Neg logit contributions | |
hotel | -0.238 |
playground | -0.235 |
city | -0.233 |
classroom | -0.232 |
street | -0.230 |
Token | He |
Token ID | 679 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
es | +0.032 |
icked | +0.028 |
on | +0.028 |
together | +0.028 |
there | +0.027 |
Neg logit contributions | |
classroom | -0.038 |
city | -0.037 |
playground | -0.037 |
museum | -0.037 |
ey | -0.037 |
Token | went |
Token ID | 1816 |
Feature activation | -4.11e-02 |
Pos logit contributions | |
es | +0.024 |
on | +0.021 |
icked | +0.021 |
together | +0.021 |
there | +0.021 |
Neg logit contributions | |
hotel | -0.040 |
ey | -0.038 |
city | -0.038 |
classroom | -0.038 |
principal | -0.038 |
Token | outside |
Token ID | 2354 |
Feature activation | -6.27e-02 |
Pos logit contributions | |
es | +0.046 |
on | +0.041 |
together | +0.040 |
icked | +0.039 |
there | +0.039 |
Neg logit contributions | |
hotel | -0.106 |
ey | -0.104 |
city | -0.102 |
classroom | -0.102 |
principal | -0.102 |
Token | and |
Token ID | 290 |
Feature activation | -3.33e-02 |
Pos logit contributions | |
es | +0.068 |
icked | +0.065 |
on | +0.046 |
there | +0.046 |
together | +0.044 |
Neg logit contributions | |
hotel | -0.167 |
classroom | -0.162 |
city | -0.162 |
playground | -0.159 |
museum | -0.157 |
Token | the |
Token ID | 262 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
former | +0.084 |
ey | +0.081 |
hotel | +0.081 |
doorway | +0.080 |
principal | +0.080 |
Neg logit contributions | |
there | -0.022 |
on | -0.019 |
es | -0.019 |
it | -0.018 |
together | -0.017 |
Token | desks |
Token ID | 44301 |
Feature activation | -6.46e-02 |
Pos logit contributions | |
former | +0.197 |
ostr | +0.177 |
ey | +0.173 |
ric | +0.173 |
priest | +0.169 |
Neg logit contributions | |
on | -0.200 |
there | -0.191 |
with | -0.189 |
at | -0.182 |
in | -0.182 |
Token | were |
Token ID | 547 |
Feature activation | -1.00e-02 |
Pos logit contributions | |
es | +0.090 |
icked | +0.085 |
together | +0.076 |
on | +0.074 |
against | +0.072 |
Neg logit contributions | |
hotel | -0.144 |
classroom | -0.142 |
city | -0.142 |
museum | -0.139 |
playground | -0.137 |
Token | in |
Token ID | 287 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
es | +0.016 |
on | +0.015 |
together | +0.014 |
there | +0.014 |
icked | +0.014 |
Neg logit contributions | |
hotel | -0.021 |
ey | -0.021 |
city | -0.020 |
principal | -0.020 |
classroom | -0.020 |
Token | dis |
Token ID | 595 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
there | +0.013 |
on | +0.013 |
es | +0.013 |
together | +0.012 |
in | +0.011 |
Neg logit contributions | |
ey | -0.029 |
hotel | -0.028 |
principal | -0.028 |
orchestra | -0.027 |
priest | -0.027 |
Token | array |
Token ID | 18747 |
Feature activation | -9.39e-02 |
Pos logit contributions | |
hotel | +0.021 |
ey | +0.020 |
city | +0.020 |
classroom | +0.020 |
principal | +0.019 |
Neg logit contributions | |
es | -0.027 |
on | -0.025 |
icked | -0.025 |
together | -0.025 |
there | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
icked | +0.112 |
es | +0.104 |
together | +0.076 |
iled | +0.075 |
there | +0.074 |
Neg logit contributions | |
hotel | -0.232 |
city | -0.226 |
playground | -0.226 |
classroom | -0.225 |
beach | -0.224 |
Token | papers |
Token ID | 9473 |
Feature activation | -6.31e-02 |
Pos logit contributions | |
es | +0.035 |
on | +0.033 |
together | +0.033 |
there | +0.033 |
icked | +0.032 |
Neg logit contributions | |
hotel | -0.044 |
ey | -0.043 |
principal | -0.041 |
city | -0.041 |
classroom | -0.041 |
Token | were |
Token ID | 547 |
Feature activation | -6.47e-03 |
Pos logit contributions | |
es | +0.080 |
icked | +0.073 |
together | +0.064 |
on | +0.062 |
against | +0.060 |
Neg logit contributions | |
hotel | -0.144 |
classroom | -0.142 |
city | -0.140 |
playground | -0.140 |
museum | -0.138 |
Token | scattered |
Token ID | 16830 |
Feature activation | -6.26e-02 |
Pos logit contributions | |
es | +0.012 |
icked | +0.011 |
together | +0.011 |
on | +0.010 |
there | +0.010 |
Neg logit contributions | |
hotel | -0.012 |
classroom | -0.011 |
city | -0.011 |
principal | -0.011 |
ey | -0.011 |
Token | everywhere |
Token ID | 8347 |
Feature activation | -5.48e-02 |
Pos logit contributions | |
es | +0.083 |
icked | +0.076 |
together | +0.060 |
on | +0.059 |
there | +0.058 |
Neg logit contributions | |
hotel | -0.147 |
city | -0.144 |
classroom | -0.142 |
museum | -0.140 |
mall | -0.138 |
Token | . |
Token ID | 13 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
es | +0.032 |
icked | +0.030 |
on | +0.025 |
together | +0.023 |
there | +0.022 |
Neg logit contributions | |
hotel | -0.106 |
classroom | -0.104 |
city | -0.104 |
playground | -0.103 |
ey | -0.102 |
Token | She |
Token ID | 1375 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
es | +0.032 |
on | +0.028 |
icked | +0.028 |
together | +0.027 |
there | +0.027 |
Neg logit contributions | |
classroom | -0.046 |
hotel | -0.046 |
ey | -0.046 |
city | -0.045 |
playground | -0.045 |
Token | dug |
Token ID | 18735 |
Feature activation | -6.41e-02 |
Pos logit contributions | |
es | +0.021 |
on | +0.019 |
icked | +0.019 |
together | +0.018 |
there | +0.018 |
Neg logit contributions | |
hotel | -0.028 |
ey | -0.027 |
classroom | -0.027 |
city | -0.027 |
playground | -0.026 |
Token | a |
Token ID | 257 |
Feature activation | +5.28e-02 |
Pos logit contributions | |
es | +0.079 |
icked | +0.071 |
together | +0.060 |
there | +0.059 |
on | +0.058 |
Neg logit contributions | |
hotel | -0.151 |
ey | -0.150 |
city | -0.145 |
playground | -0.144 |
beach | -0.143 |
Token | little |
Token ID | 1310 |
Feature activation | +4.45e-02 |
Pos logit contributions | |
principal | +0.102 |
carn | +0.092 |
opl | +0.091 |
hotel | +0.091 |
mall | +0.090 |
Neg logit contributions | |
on | -0.102 |
together | -0.100 |
there | -0.098 |
in | -0.096 |
with | -0.095 |
Token | hole |
Token ID | 7604 |
Feature activation | -9.36e-02 |
Pos logit contributions | |
principal | +0.073 |
hotel | +0.072 |
classroom | +0.068 |
ey | +0.068 |
city | +0.068 |
Neg logit contributions | |
es | -0.090 |
on | -0.088 |
together | -0.088 |
there | -0.086 |
in | -0.081 |
Token | and |
Token ID | 290 |
Feature activation | -2.83e-02 |
Pos logit contributions | |
es | +0.101 |
icked | +0.090 |
hot | +0.067 |
rums | +0.067 |
iled | +0.061 |
Neg logit contributions | |
doorway | -0.217 |
hotel | -0.211 |
ey | -0.211 |
playground | -0.207 |
beach | -0.204 |
Token | carefully |
Token ID | 7773 |
Feature activation | -4.34e-02 |
Pos logit contributions | |
es | +0.044 |
icked | +0.039 |
on | +0.038 |
there | +0.036 |
together | +0.036 |
Neg logit contributions | |
hotel | -0.058 |
playground | -0.056 |
ey | -0.056 |
classroom | -0.056 |
beach | -0.055 |
Token | placed |
Token ID | 4624 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
es | +0.081 |
icked | +0.070 |
together | +0.066 |
on | +0.066 |
there | +0.064 |
Neg logit contributions | |
beach | -0.072 |
playground | -0.072 |
ey | -0.070 |
hotel | -0.070 |
doorway | -0.070 |
Token | the |
Token ID | 262 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
es | +0.033 |
icked | +0.026 |
together | +0.022 |
on | +0.019 |
there | +0.018 |
Neg logit contributions | |
ey | -0.107 |
hotel | -0.106 |
city | -0.104 |
museum | -0.103 |
classroom | -0.103 |
Token | flower |
Token ID | 15061 |
Feature activation | -3.22e-02 |
Pos logit contributions | |
principal | +0.224 |
hotel | +0.222 |
carn | +0.207 |
ey | +0.206 |
classroom | +0.205 |
Neg logit contributions | |
on | -0.336 |
there | -0.329 |
es | -0.329 |
together | -0.316 |
in | -0.313 |
Token | pirates |
Token ID | 27516 |
Feature activation | -7.58e-02 |
Pos logit contributions | |
icked | +0.048 |
es | +0.047 |
on | +0.034 |
together | +0.033 |
there | +0.031 |
Neg logit contributions | |
hotel | -0.106 |
classroom | -0.103 |
museum | -0.103 |
city | -0.102 |
playground | -0.102 |
Token | . |
Token ID | 13 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
icked | +0.084 |
es | +0.071 |
together | +0.048 |
iled | +0.047 |
there | +0.047 |
Neg logit contributions | |
hotel | -0.196 |
city | -0.189 |
playground | -0.189 |
classroom | -0.189 |
beach | -0.188 |
Token | They |
Token ID | 1119 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
es | +0.036 |
on | +0.031 |
icked | +0.031 |
together | +0.030 |
there | +0.029 |
Neg logit contributions | |
hotel | -0.054 |
classroom | -0.053 |
city | -0.053 |
playground | -0.052 |
museum | -0.052 |
Token | put |
Token ID | 1234 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
es | +0.017 |
on | +0.017 |
there | +0.016 |
together | +0.016 |
icked | +0.015 |
Neg logit contributions | |
ey | -0.022 |
hotel | -0.022 |
principal | -0.022 |
former | -0.021 |
classroom | -0.021 |
Token | on |
Token ID | 319 |
Feature activation | -3.89e-02 |
Pos logit contributions | |
es | +0.033 |
icked | +0.028 |
together | +0.026 |
on | +0.026 |
there | +0.025 |
Neg logit contributions | |
hotel | -0.082 |
ey | -0.080 |
city | -0.080 |
classroom | -0.079 |
museum | -0.079 |
Token | hats |
Token ID | 23910 |
Feature activation | -9.35e-02 |
Pos logit contributions | |
es | +0.038 |
on | +0.038 |
together | +0.038 |
there | +0.036 |
in | +0.031 |
Neg logit contributions | |
principal | -0.104 |
ey | -0.104 |
hotel | -0.103 |
priest | -0.103 |
orchestra | -0.102 |
Token | and |
Token ID | 290 |
Feature activation | -7.39e-03 |
Pos logit contributions | |
icked | +0.121 |
es | +0.116 |
able | +0.085 |
there | +0.083 |
together | +0.081 |
Neg logit contributions | |
hotel | -0.204 |
city | -0.200 |
airport | -0.200 |
classroom | -0.199 |
playground | -0.199 |
Token | scar |
Token ID | 10153 |
Feature activation | -6.14e-02 |
Pos logit contributions | |
es | +0.015 |
on | +0.014 |
together | +0.014 |
there | +0.014 |
icked | +0.014 |
Neg logit contributions | |
hotel | -0.012 |
ey | -0.012 |
principal | -0.012 |
city | -0.012 |
classroom | -0.012 |
Token | ves |
Token ID | 1158 |
Feature activation | -7.12e-02 |
Pos logit contributions | |
icked | +0.093 |
there | +0.092 |
es | +0.088 |
instead | +0.087 |
together | +0.083 |
Neg logit contributions | |
hotel | -0.101 |
loudspe | -0.098 |
principal | -0.097 |
tant | -0.096 |
orchestra | -0.096 |
Token | and |
Token ID | 290 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
icked | +0.083 |
es | +0.082 |
able | +0.065 |
there | +0.062 |
pped | +0.060 |
Neg logit contributions | |
bus | -0.141 |
tour | -0.141 |
city | -0.141 |
hotel | -0.140 |
playground | -0.140 |
Token | pretend |
Token ID | 16614 |
Feature activation | -5.20e-02 |
Pos logit contributions | |
es | +0.035 |
on | +0.034 |
there | +0.033 |
together | +0.033 |
icked | +0.032 |
Neg logit contributions | |
hotel | -0.034 |
ey | -0.033 |
principal | -0.032 |
city | -0.032 |
classroom | -0.032 |
Token | car |
Token ID | 1097 |
Feature activation | -4.89e-02 |
Pos logit contributions | |
ey | +0.196 |
former | +0.193 |
principal | +0.192 |
Ta | +0.187 |
orchestra | +0.187 |
Neg logit contributions | |
on | -0.285 |
there | -0.276 |
together | -0.265 |
in | -0.258 |
at | -0.247 |
Token | . |
Token ID | 13 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
es | +0.049 |
icked | +0.046 |
there | +0.038 |
on | +0.037 |
together | +0.036 |
Neg logit contributions | |
hotel | -0.128 |
city | -0.126 |
classroom | -0.124 |
museum | -0.124 |
playground | -0.123 |
Token | They |
Token ID | 1119 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
es | +0.022 |
on | +0.019 |
together | +0.019 |
icked | +0.019 |
there | +0.018 |
Neg logit contributions | |
hotel | -0.028 |
ey | -0.027 |
principal | -0.027 |
classroom | -0.027 |
city | -0.027 |
Token | wear |
Token ID | 5806 |
Feature activation | -4.91e-02 |
Pos logit contributions | |
on | +0.023 |
es | +0.023 |
there | +0.023 |
together | +0.021 |
in | +0.021 |
Neg logit contributions | |
ey | -0.028 |
principal | -0.028 |
former | -0.028 |
igsaw | -0.027 |
classroom | -0.027 |
Token | big |
Token ID | 1263 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
es | +0.077 |
on | +0.070 |
icked | +0.069 |
together | +0.069 |
there | +0.068 |
Neg logit contributions | |
hotel | -0.104 |
ey | -0.101 |
city | -0.099 |
classroom | -0.099 |
principal | -0.099 |
Token | hats |
Token ID | 23910 |
Feature activation | -9.34e-02 |
Pos logit contributions | |
hotel | +0.021 |
principal | +0.020 |
former | +0.020 |
ey | +0.020 |
classroom | +0.020 |
Neg logit contributions | |
es | -0.029 |
on | -0.028 |
there | -0.027 |
together | -0.027 |
icked | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
icked | +0.125 |
es | +0.123 |
able | +0.088 |
rums | +0.087 |
there | +0.086 |
Neg logit contributions | |
hotel | -0.192 |
airport | -0.189 |
bus | -0.189 |
city | -0.188 |
playground | -0.188 |
Token | coats |
Token ID | 30720 |
Feature activation | -9.45e-02 |
Pos logit contributions | |
es | +0.026 |
on | +0.025 |
together | +0.024 |
there | +0.024 |
icked | +0.024 |
Neg logit contributions | |
hotel | -0.023 |
ey | -0.023 |
principal | -0.022 |
city | -0.022 |
classroom | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
es | +0.115 |
icked | +0.107 |
there | +0.076 |
able | +0.076 |
together | +0.071 |
Neg logit contributions | |
hotel | -0.220 |
bus | -0.217 |
playground | -0.216 |
city | -0.215 |
classroom | -0.214 |
Token | They |
Token ID | 1119 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
es | +0.019 |
icked | +0.016 |
on | +0.016 |
together | +0.016 |
there | +0.015 |
Neg logit contributions | |
hotel | -0.023 |
classroom | -0.022 |
city | -0.022 |
principal | -0.022 |
ey | -0.021 |
Token | have |
Token ID | 423 |
Feature activation | -9.70e-03 |
Pos logit contributions | |
on | +0.029 |
es | +0.027 |
there | +0.027 |
together | +0.026 |
in | +0.026 |
Neg logit contributions | |
ey | -0.033 |
former | -0.032 |
principal | -0.032 |
doorway | -0.032 |
igsaw | -0.032 |
Token | She |
Token ID | 1375 |
Feature activation | -3.93e-03 |
Pos logit contributions | |
es | +0.013 |
on | +0.011 |
icked | +0.011 |
together | +0.011 |
there | +0.011 |
Neg logit contributions | |
hotel | -0.023 |
ey | -0.022 |
city | -0.022 |
classroom | -0.022 |
principal | -0.022 |
Token | put |
Token ID | 1234 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
es | +0.006 |
on | +0.005 |
together | +0.005 |
there | +0.005 |
icked | +0.005 |
Neg logit contributions | |
hotel | -0.009 |
ey | -0.009 |
principal | -0.008 |
city | -0.008 |
classroom | -0.008 |
Token | on |
Token ID | 319 |
Feature activation | -3.27e-02 |
Pos logit contributions | |
es | +0.014 |
icked | +0.008 |
together | +0.007 |
instead | +0.006 |
there | +0.005 |
Neg logit contributions | |
ey | -0.075 |
hotel | -0.075 |
classroom | -0.073 |
city | -0.072 |
mirror | -0.072 |
Token | her |
Token ID | 607 |
Feature activation | +8.07e-02 |
Pos logit contributions | |
on | +0.030 |
es | +0.030 |
there | +0.028 |
together | +0.028 |
in | +0.024 |
Neg logit contributions | |
principal | -0.089 |
hotel | -0.089 |
ey | -0.087 |
museum | -0.086 |
priest | -0.086 |
Token | rain |
Token ID | 6290 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
principal | +0.140 |
hotel | +0.137 |
ey | +0.136 |
museum | +0.133 |
priest | +0.131 |
Neg logit contributions | |
on | -0.158 |
there | -0.152 |
es | -0.148 |
together | -0.146 |
in | -0.143 |
Token | coat |
Token ID | 31434 |
Feature activation | -9.32e-02 |
Pos logit contributions | |
es | +0.071 |
icked | +0.066 |
on | +0.064 |
there | +0.063 |
together | +0.063 |
Neg logit contributions | |
hotel | -0.068 |
classroom | -0.065 |
city | -0.064 |
ey | -0.064 |
playground | -0.064 |
Token | and |
Token ID | 290 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
es | +0.107 |
icked | +0.100 |
there | +0.078 |
together | +0.076 |
on | +0.070 |
Neg logit contributions | |
hotel | -0.232 |
playground | -0.228 |
city | -0.226 |
classroom | -0.225 |
airport | -0.222 |
Token | got |
Token ID | 1392 |
Feature activation | -3.18e-02 |
Pos logit contributions | |
es | +0.041 |
on | +0.038 |
icked | +0.037 |
together | +0.037 |
there | +0.037 |
Neg logit contributions | |
hotel | -0.041 |
ey | -0.039 |
city | -0.039 |
classroom | -0.039 |
principal | -0.038 |
Token | her |
Token ID | 607 |
Feature activation | +6.93e-02 |
Pos logit contributions | |
es | +0.043 |
icked | +0.036 |
on | +0.033 |
together | +0.033 |
there | +0.032 |
Neg logit contributions | |
hotel | -0.076 |
ey | -0.074 |
classroom | -0.074 |
city | -0.073 |
playground | -0.072 |
Token | toy |
Token ID | 13373 |
Feature activation | +8.47e-03 |
Pos logit contributions | |
principal | +0.114 |
ey | +0.110 |
hotel | +0.109 |
museum | +0.107 |
city | +0.106 |
Neg logit contributions | |
on | -0.145 |
there | -0.140 |
together | -0.137 |
es | -0.134 |
in | -0.133 |
Token | boat |
Token ID | 8848 |
Feature activation | -5.89e-02 |
Pos logit contributions | |
hotel | +0.013 |
city | +0.012 |
classroom | +0.012 |
playground | +0.012 |
mall | +0.012 |
Neg logit contributions | |
es | -0.019 |
icked | -0.017 |
there | -0.016 |
together | -0.016 |
on | -0.016 |
Token | of |
Token ID | 286 |
Feature activation | +4.40e-02 |
Pos logit contributions | |
es | +0.035 |
on | +0.030 |
icked | +0.030 |
together | +0.030 |
there | +0.029 |
Neg logit contributions | |
hotel | -0.070 |
ey | -0.069 |
city | -0.067 |
classroom | -0.067 |
museum | -0.066 |
Token | dresses |
Token ID | 27309 |
Feature activation | -7.82e-02 |
Pos logit contributions | |
principal | +0.076 |
hotel | +0.070 |
courtroom | +0.070 |
former | +0.070 |
city | +0.070 |
Neg logit contributions | |
on | -0.090 |
there | -0.087 |
es | -0.086 |
together | -0.084 |
in | -0.084 |
Token | and |
Token ID | 290 |
Feature activation | -7.42e-03 |
Pos logit contributions | |
icked | +0.075 |
es | +0.072 |
iled | +0.050 |
there | +0.049 |
together | +0.049 |
Neg logit contributions | |
hotel | -0.191 |
classroom | -0.187 |
playground | -0.185 |
beach | -0.184 |
city | -0.184 |
Token | shoes |
Token ID | 10012 |
Feature activation | -8.34e-02 |
Pos logit contributions | |
es | +0.016 |
on | +0.015 |
together | +0.015 |
there | +0.015 |
in | +0.014 |
Neg logit contributions | |
hotel | -0.012 |
principal | -0.012 |
ey | -0.012 |
city | -0.011 |
courtroom | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
es | +0.081 |
icked | +0.076 |
there | +0.054 |
together | +0.052 |
iled | +0.048 |
Neg logit contributions | |
hotel | -0.213 |
classroom | -0.204 |
mall | -0.204 |
salon | -0.204 |
beach | -0.203 |
Token | hats |
Token ID | 23910 |
Feature activation | -9.30e-02 |
Pos logit contributions | |
on | +0.021 |
es | +0.021 |
there | +0.020 |
together | +0.020 |
in | +0.019 |
Neg logit contributions | |
hotel | -0.019 |
principal | -0.018 |
ey | -0.018 |
city | -0.017 |
courtroom | -0.017 |
Token | to |
Token ID | 284 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
es | +0.095 |
icked | +0.089 |
iled | +0.073 |
there | +0.072 |
together | +0.066 |
Neg logit contributions | |
hotel | -0.211 |
playground | -0.205 |
beach | -0.204 |
bus | -0.204 |
classroom | -0.204 |
Token | choose |
Token ID | 3853 |
Feature activation | -3.96e-02 |
Pos logit contributions | |
ey | +0.034 |
hotel | +0.034 |
principal | +0.033 |
doorway | +0.033 |
city | +0.033 |
Neg logit contributions | |
on | -0.022 |
es | -0.021 |
together | -0.020 |
there | -0.020 |
in | -0.019 |
Token | from |
Token ID | 422 |
Feature activation | +3.66e-03 |
Pos logit contributions | |
es | +0.044 |
icked | +0.040 |
on | +0.034 |
together | +0.034 |
there | +0.034 |
Neg logit contributions | |
hotel | -0.099 |
classroom | -0.096 |
city | -0.095 |
museum | -0.095 |
ey | -0.094 |
Token | . |
Token ID | 13 |
Feature activation | -9.84e-03 |
Pos logit contributions | |
hotel | +0.010 |
city | +0.010 |
classroom | +0.010 |
museum | +0.010 |
principal | +0.009 |
Neg logit contributions | |
es | -0.004 |
icked | -0.003 |
together | -0.003 |
on | -0.003 |
there | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -6.84e-03 |
Pos logit contributions | |
es | +0.014 |
on | +0.012 |
icked | +0.012 |
together | +0.012 |
there | +0.011 |
Neg logit contributions | |
hotel | -0.023 |
ey | -0.022 |
city | -0.022 |
classroom | -0.022 |
playground | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | -3.39e-02 |
Pos logit contributions | |
icked | +0.030 |
es | +0.029 |
together | +0.023 |
able | +0.021 |
pped | +0.020 |
Neg logit contributions | |
hotel | -0.070 |
city | -0.066 |
ostr | -0.064 |
mall | -0.064 |
priest | -0.064 |
Token | she |
Token ID | 673 |
Feature activation | -8.59e-03 |
Pos logit contributions | |
icked | +0.049 |
es | +0.048 |
together | +0.041 |
on | +0.038 |
iled | +0.037 |
Neg logit contributions | |
hotel | -0.076 |
city | -0.074 |
ostr | -0.071 |
priest | -0.071 |
zoo | -0.070 |
Token | would |
Token ID | 561 |
Feature activation | -2.75e-02 |
Pos logit contributions | |
es | +0.012 |
on | +0.011 |
together | +0.010 |
icked | +0.010 |
there | +0.010 |
Neg logit contributions | |
hotel | -0.020 |
ey | -0.020 |
principal | -0.019 |
city | -0.019 |
classroom | -0.019 |
Token | walk |
Token ID | 2513 |
Feature activation | -6.22e-02 |
Pos logit contributions | |
on | +0.046 |
es | +0.046 |
there | +0.045 |
together | +0.044 |
in | +0.043 |
Neg logit contributions | |
hotel | -0.052 |
principal | -0.052 |
ey | -0.051 |
classroom | -0.050 |
city | -0.050 |
Token | along |
Token ID | 1863 |
Feature activation | -2.49e-02 |
Pos logit contributions | |
es | +0.067 |
icked | +0.063 |
there | +0.047 |
together | +0.047 |
on | +0.046 |
Neg logit contributions | |
hotel | -0.162 |
city | -0.157 |
classroom | -0.157 |
playground | -0.154 |
beach | -0.153 |
Token | the |
Token ID | 262 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
es | +0.022 |
on | +0.020 |
together | +0.019 |
there | +0.018 |
icked | +0.017 |
Neg logit contributions | |
hotel | -0.070 |
ey | -0.069 |
principal | -0.067 |
city | -0.067 |
classroom | -0.067 |
Token | beach |
Token ID | 10481 |
Feature activation | -4.99e-02 |
Pos logit contributions | |
principal | +0.188 |
babys | +0.186 |
former | +0.181 |
igsaw | +0.179 |
Ta | +0.176 |
Neg logit contributions | |
on | -0.309 |
there | -0.304 |
in | -0.296 |
it | -0.286 |
together | -0.273 |
Token | , |
Token ID | 11 |
Feature activation | -3.09e-02 |
Pos logit contributions | |
es | +0.046 |
icked | +0.042 |
on | +0.037 |
together | +0.035 |
there | +0.035 |
Neg logit contributions | |
hotel | -0.138 |
city | -0.135 |
classroom | -0.134 |
playground | -0.132 |
ey | -0.132 |
Token | looking |
Token ID | 2045 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
es | +0.041 |
on | +0.039 |
together | +0.038 |
there | +0.037 |
icked | +0.036 |
Neg logit contributions | |
hotel | -0.073 |
ey | -0.071 |
principal | -0.070 |
city | -0.069 |
classroom | -0.069 |
Token | for |
Token ID | 329 |
Feature activation | +5.96e-03 |
Pos logit contributions | |
es | +0.028 |
icked | +0.023 |
on | +0.020 |
together | +0.020 |
there | +0.020 |
Neg logit contributions | |
hotel | -0.095 |
ey | -0.093 |
city | -0.092 |
classroom | -0.092 |
museum | -0.090 |
Token | shells |
Token ID | 19679 |
Feature activation | -5.11e-02 |
Pos logit contributions | |
hotel | +0.011 |
principal | +0.011 |
ey | +0.011 |
classroom | +0.010 |
city | +0.010 |
Neg logit contributions | |
es | -0.011 |
on | -0.011 |
there | -0.010 |
together | -0.010 |
icked | -0.010 |
Token | happy |
Token ID | 3772 |
Feature activation | -2.97e-02 |
Pos logit contributions | |
hotel | +0.015 |
ey | +0.015 |
city | +0.014 |
classroom | +0.014 |
principal | +0.014 |
Neg logit contributions | |
es | -0.021 |
on | -0.020 |
icked | -0.019 |
together | -0.019 |
there | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | -1.65e-02 |
Pos logit contributions | |
icked | +0.045 |
es | +0.040 |
pped | +0.033 |
hot | +0.032 |
bled | +0.030 |
Neg logit contributions | |
beach | -0.061 |
playground | -0.061 |
hotel | -0.059 |
classroom | -0.058 |
zoo | -0.058 |
Token | grateful |
Token ID | 14066 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
es | +0.027 |
icked | +0.025 |
on | +0.024 |
together | +0.023 |
there | +0.023 |
Neg logit contributions | |
hotel | -0.033 |
ey | -0.031 |
classroom | -0.031 |
playground | -0.031 |
city | -0.031 |
Token | to |
Token ID | 284 |
Feature activation | -5.34e-03 |
Pos logit contributions | |
es | +0.024 |
icked | +0.020 |
pped | +0.016 |
together | +0.016 |
there | +0.016 |
Neg logit contributions | |
hotel | -0.050 |
talent | -0.048 |
classroom | -0.047 |
beach | -0.047 |
museum | -0.047 |
Token | the |
Token ID | 262 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
es | +0.006 |
on | +0.006 |
together | +0.005 |
there | +0.005 |
icked | +0.005 |
Neg logit contributions | |
ey | -0.014 |
hotel | -0.014 |
city | -0.013 |
museum | -0.013 |
playground | -0.013 |
Token | little |
Token ID | 1310 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
ey | +0.240 |
opl | +0.230 |
principal | +0.224 |
former | +0.214 |
igsaw | +0.214 |
Neg logit contributions | |
there | -0.285 |
on | -0.274 |
in | -0.257 |
with | -0.255 |
it | -0.253 |
Token | girl |
Token ID | 2576 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
ey | +0.141 |
principal | +0.139 |
classroom | +0.132 |
city | +0.129 |
hotel | +0.128 |
Neg logit contributions | |
there | -0.276 |
on | -0.274 |
es | -0.263 |
together | -0.260 |
in | -0.256 |
Token | and |
Token ID | 290 |
Feature activation | -7.02e-03 |
Pos logit contributions | |
es | +0.013 |
icked | +0.013 |
there | +0.011 |
together | +0.010 |
on | +0.010 |
Neg logit contributions | |
hotel | -0.036 |
classroom | -0.035 |
city | -0.035 |
principal | -0.034 |
playground | -0.034 |
Token | her |
Token ID | 607 |
Feature activation | +8.42e-02 |
Pos logit contributions | |
es | +0.008 |
on | +0.008 |
together | +0.007 |
icked | +0.007 |
there | +0.007 |
Neg logit contributions | |
hotel | -0.018 |
ey | -0.017 |
city | -0.017 |
classroom | -0.017 |
principal | -0.017 |
Token | family |
Token ID | 1641 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
ey | +0.160 |
Ta | +0.153 |
orchestra | +0.151 |
principal | +0.145 |
hotel | +0.145 |
Neg logit contributions | |
on | -0.156 |
there | -0.150 |
together | -0.144 |
in | -0.142 |
es | -0.142 |
Token | . |
Token ID | 13 |
Feature activation | -3.37e-03 |
Pos logit contributions | |
es | +0.015 |
icked | +0.013 |
there | +0.011 |
on | +0.011 |
together | +0.010 |
Neg logit contributions | |
hotel | -0.044 |
city | -0.043 |
playground | -0.042 |
classroom | -0.042 |
museum | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
hotel | +0.061 |
ey | +0.060 |
city | +0.059 |
classroom | +0.059 |
principal | +0.059 |
Neg logit contributions | |
es | -0.039 |
on | -0.035 |
icked | -0.034 |
together | -0.034 |
there | -0.034 |
Token | snow |
Token ID | 6729 |
Feature activation | +3.23e-03 |
Pos logit contributions | |
ey | +0.214 |
principal | +0.210 |
hotel | +0.207 |
classroom | +0.205 |
courtroom | +0.202 |
Neg logit contributions | |
on | -0.260 |
there | -0.258 |
es | -0.257 |
together | -0.243 |
icked | -0.236 |
Token | man |
Token ID | 805 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
hotel | +0.004 |
city | +0.004 |
ostr | +0.004 |
zoo | +0.004 |
tour | +0.004 |
Neg logit contributions | |
es | -0.007 |
icked | -0.007 |
there | -0.007 |
on | -0.007 |
together | -0.007 |
Token | went |
Token ID | 1816 |
Feature activation | -3.40e-02 |
Pos logit contributions | |
es | +0.049 |
icked | +0.044 |
there | +0.043 |
on | +0.042 |
together | +0.042 |
Neg logit contributions | |
hotel | -0.071 |
city | -0.069 |
airport | -0.066 |
tour | -0.066 |
ostr | -0.065 |
Token | to |
Token ID | 284 |
Feature activation | -6.28e-03 |
Pos logit contributions | |
es | +0.040 |
icked | +0.037 |
instead | +0.028 |
there | +0.026 |
though | +0.025 |
Neg logit contributions | |
hotel | -0.087 |
city | -0.083 |
tour | -0.082 |
playground | -0.082 |
zoo | -0.082 |
Token | the |
Token ID | 262 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
on | +0.005 |
together | +0.005 |
there | +0.005 |
es | +0.005 |
in | +0.004 |
Neg logit contributions | |
ey | -0.018 |
principal | -0.017 |
orchestra | -0.017 |
hotel | -0.017 |
classroom | -0.017 |
Token | barn |
Token ID | 25203 |
Feature activation | -5.07e-02 |
Pos logit contributions | |
ey | +0.194 |
igsaw | +0.183 |
principal | +0.177 |
orchestra | +0.173 |
tant | +0.171 |
Neg logit contributions | |
on | -0.256 |
there | -0.253 |
together | -0.242 |
es | -0.239 |
in | -0.234 |
Token | . |
Token ID | 13 |
Feature activation | +1.79e-03 |
Pos logit contributions | |
es | +0.050 |
icked | +0.045 |
on | +0.041 |
there | +0.039 |
together | +0.038 |
Neg logit contributions | |
hotel | -0.137 |
city | -0.133 |
classroom | -0.131 |
playground | -0.131 |
museum | -0.130 |
Token | The |
Token ID | 383 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
hotel | +0.004 |
city | +0.004 |
museum | +0.004 |
zoo | +0.004 |
former | +0.004 |
Neg logit contributions | |
es | -0.003 |
icked | -0.002 |
on | -0.002 |
there | -0.002 |
together | -0.002 |
Token | snow |
Token ID | 6729 |
Feature activation | -6.96e-03 |
Pos logit contributions | |
ey | +0.222 |
principal | +0.220 |
igsaw | +0.219 |
courtroom | +0.218 |
classroom | +0.210 |
Neg logit contributions | |
on | -0.238 |
there | -0.236 |
es | -0.223 |
together | -0.219 |
in | -0.213 |
Token | man |
Token ID | 805 |
Feature activation | -4.01e-02 |
Pos logit contributions | |
es | +0.016 |
icked | +0.016 |
together | +0.015 |
on | +0.015 |
there | +0.015 |
Neg logit contributions | |
hotel | -0.009 |
city | -0.008 |
playground | -0.008 |
ostr | -0.008 |
zoo | -0.008 |
Token | The |
Token ID | 383 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
es | +0.025 |
icked | +0.022 |
on | +0.022 |
together | +0.022 |
there | +0.021 |
Neg logit contributions | |
hotel | -0.037 |
ey | -0.036 |
principal | -0.036 |
city | -0.035 |
classroom | -0.035 |
Token | cow |
Token ID | 9875 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
ey | +0.243 |
hotel | +0.233 |
principal | +0.225 |
former | +0.224 |
classroom | +0.223 |
Neg logit contributions | |
on | -0.244 |
es | -0.239 |
there | -0.238 |
together | -0.222 |
in | -0.217 |
Token | felt |
Token ID | 2936 |
Feature activation | -8.18e-03 |
Pos logit contributions | |
es | +0.048 |
icked | +0.044 |
on | +0.043 |
together | +0.043 |
there | +0.042 |
Neg logit contributions | |
hotel | -0.073 |
city | -0.070 |
ey | -0.070 |
principal | -0.070 |
classroom | -0.069 |
Token | sad |
Token ID | 6507 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
es | +0.015 |
on | +0.014 |
icked | +0.014 |
together | +0.014 |
there | +0.014 |
Neg logit contributions | |
hotel | -0.015 |
ey | -0.014 |
city | -0.014 |
classroom | -0.014 |
principal | -0.014 |
Token | for |
Token ID | 329 |
Feature activation | -2.80e-03 |
Pos logit contributions | |
icked | +0.035 |
es | +0.035 |
pped | +0.026 |
iled | +0.025 |
there | +0.024 |
Neg logit contributions | |
hotel | -0.046 |
beach | -0.043 |
classroom | -0.043 |
mirror | -0.042 |
city | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
es | +0.004 |
icked | +0.004 |
together | +0.003 |
on | +0.003 |
there | +0.003 |
Neg logit contributions | |
hotel | -0.006 |
principal | -0.006 |
museum | -0.006 |
city | -0.006 |
ey | -0.006 |
Token | farmer |
Token ID | 18739 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
ey | +0.248 |
hotel | +0.236 |
principal | +0.225 |
classroom | +0.222 |
city | +0.219 |
Neg logit contributions | |
on | -0.288 |
there | -0.283 |
es | -0.276 |
together | -0.270 |
in | -0.260 |
Token | and |
Token ID | 290 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
es | +0.021 |
icked | +0.020 |
together | +0.018 |
on | +0.017 |
there | +0.017 |
Neg logit contributions | |
hotel | -0.049 |
city | -0.048 |
principal | -0.048 |
classroom | -0.047 |
museum | -0.047 |
Token | m |
Token ID | 285 |
Feature activation | +7.43e-02 |
Pos logit contributions | |
es | +0.020 |
icked | +0.017 |
on | +0.017 |
together | +0.017 |
there | +0.015 |
Neg logit contributions | |
hotel | -0.030 |
principal | -0.028 |
classroom | -0.028 |
city | -0.028 |
playground | -0.028 |
Token | oo |
Token ID | 2238 |
Feature activation | -2.97e-02 |
Pos logit contributions | |
hotel | +0.143 |
ey | +0.139 |
principal | +0.137 |
city | +0.137 |
classroom | +0.135 |
Neg logit contributions | |
es | -0.128 |
on | -0.122 |
together | -0.119 |
there | -0.118 |
icked | -0.116 |
Token | ed |
Token ID | 276 |
Feature activation | -5.32e-02 |
Pos logit contributions | |
es | +0.028 |
icked | +0.026 |
together | +0.026 |
there | +0.025 |
on | +0.024 |
Neg logit contributions | |
hotel | -0.072 |
principal | -0.069 |
ostr | -0.068 |
city | -0.067 |
babys | -0.067 |
Token | out |
Token ID | 503 |
Feature activation | -3.29e-02 |
Pos logit contributions | |
es | +0.026 |
icked | +0.023 |
on | +0.023 |
together | +0.023 |
there | +0.022 |
Neg logit contributions | |
hotel | -0.031 |
city | -0.030 |
ey | -0.029 |
playground | -0.029 |
classroom | -0.029 |
Token | sm |
Token ID | 5796 |
Feature activation | +7.20e-02 |
Pos logit contributions | |
es | +0.037 |
icked | +0.032 |
together | +0.031 |
on | +0.030 |
there | +0.029 |
Neg logit contributions | |
hotel | -0.083 |
ey | -0.080 |
city | -0.080 |
classroom | -0.080 |
principal | -0.079 |
Token | arted |
Token ID | 19112 |
Feature activation | -4.08e-02 |
Pos logit contributions | |
hotel | +0.157 |
principal | +0.152 |
classroom | +0.151 |
playground | +0.149 |
city | +0.149 |
Neg logit contributions | |
es | -0.105 |
icked | -0.102 |
on | -0.102 |
together | -0.100 |
there | -0.099 |
Token | the |
Token ID | 262 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
es | +0.046 |
icked | +0.041 |
together | +0.038 |
there | +0.037 |
on | +0.037 |
Neg logit contributions | |
hotel | -0.103 |
city | -0.101 |
ey | -0.100 |
classroom | -0.099 |
playground | -0.098 |
Token | big |
Token ID | 1263 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
ey | +0.239 |
former | +0.238 |
principal | +0.226 |
opl | +0.222 |
babys | +0.221 |
Neg logit contributions | |
on | -0.314 |
there | -0.306 |
in | -0.287 |
it | -0.275 |
with | -0.266 |
Token | scary |
Token ID | 14343 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
hotel | +0.191 |
ey | +0.187 |
principal | +0.186 |
former | +0.182 |
classroom | +0.180 |
Neg logit contributions | |
on | -0.320 |
es | -0.318 |
there | -0.312 |
together | -0.303 |
in | -0.292 |
Token | shark |
Token ID | 21027 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
hotel | +0.158 |
ey | +0.158 |
principal | +0.154 |
classroom | +0.150 |
city | +0.148 |
Neg logit contributions | |
es | -0.274 |
on | -0.268 |
there | -0.263 |
together | -0.258 |
icked | -0.250 |
Token | . |
Token ID | 13 |
Feature activation | -9.43e-03 |
Pos logit contributions | |
icked | +0.025 |
es | +0.025 |
on | +0.019 |
there | +0.019 |
together | +0.018 |
Neg logit contributions | |
hotel | -0.073 |
city | -0.070 |
playground | -0.070 |
museum | -0.069 |
classroom | -0.069 |
Token | From |
Token ID | 3574 |
Feature activation | -1.13e-03 |
Pos logit contributions | |
es | +0.015 |
on | +0.013 |
icked | +0.013 |
together | +0.013 |
there | +0.013 |
Neg logit contributions | |
hotel | -0.020 |
ey | -0.019 |
city | -0.019 |
classroom | -0.019 |
principal | -0.019 |
Token | that |
Token ID | 326 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
es | +0.001 |
icked | +0.001 |
together | +0.001 |
on | +0.001 |
against | +0.001 |
Neg logit contributions | |
hotel | -0.003 |
beach | -0.003 |
city | -0.003 |
airport | -0.003 |
museum | -0.003 |
Token | day |
Token ID | 1110 |
Feature activation | -9.37e-03 |
Pos logit contributions | |
hotel | +0.026 |
city | +0.024 |
classroom | +0.024 |
ey | +0.024 |
museum | +0.023 |
Neg logit contributions | |
es | -0.055 |
icked | -0.052 |
together | -0.050 |
on | -0.050 |
there | -0.049 |
Token | day |
Token ID | 1110 |
Feature activation | -2.42e-02 |
Pos logit contributions | |
hotel | +0.027 |
ey | +0.026 |
principal | +0.026 |
city | +0.026 |
classroom | +0.025 |
Neg logit contributions | |
es | -0.047 |
on | -0.045 |
together | -0.044 |
icked | -0.044 |
there | -0.044 |
Token | , |
Token ID | 11 |
Feature activation | -3.39e-02 |
Pos logit contributions | |
es | +0.040 |
icked | +0.032 |
able | +0.031 |
ppers | +0.029 |
perched | +0.028 |
Neg logit contributions | |
museum | -0.044 |
hotel | -0.043 |
city | -0.042 |
parade | -0.042 |
classroom | -0.042 |
Token | Tim |
Token ID | 5045 |
Feature activation | -6.13e-03 |
Pos logit contributions | |
es | +0.067 |
icked | +0.056 |
able | +0.045 |
owed | +0.045 |
iled | +0.044 |
Neg logit contributions | |
museum | -0.062 |
city | -0.061 |
playground | -0.060 |
parade | -0.060 |
classroom | -0.059 |
Token | went |
Token ID | 1816 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
es | +0.011 |
icked | +0.010 |
together | +0.010 |
on | +0.009 |
against | +0.009 |
Neg logit contributions | |
hotel | -0.012 |
classroom | -0.012 |
city | -0.011 |
playground | -0.011 |
museum | -0.011 |
Token | into |
Token ID | 656 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
es | +0.035 |
icked | +0.031 |
together | +0.023 |
instead | +0.021 |
there | +0.021 |
Neg logit contributions | |
classroom | -0.079 |
hotel | -0.078 |
museum | -0.077 |
playground | -0.076 |
city | -0.076 |
Token | the |
Token ID | 262 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
es | +0.018 |
on | +0.016 |
there | +0.015 |
together | +0.015 |
icked | +0.014 |
Neg logit contributions | |
hotel | -0.050 |
ey | -0.050 |
principal | -0.049 |
city | -0.049 |
classroom | -0.048 |
Token | maze |
Token ID | 31237 |
Feature activation | -5.74e-02 |
Pos logit contributions | |
priest | +0.167 |
babys | +0.154 |
orchestra | +0.150 |
pastor | +0.150 |
tant | +0.149 |
Neg logit contributions | |
there | -0.233 |
on | -0.232 |
together | -0.215 |
in | -0.210 |
at | -0.210 |
Token | to |
Token ID | 284 |
Feature activation | +9.25e-03 |
Pos logit contributions | |
es | +0.054 |
icked | +0.050 |
on | +0.038 |
there | +0.036 |
together | +0.035 |
Neg logit contributions | |
hotel | -0.157 |
city | -0.156 |
classroom | -0.156 |
museum | -0.154 |
playground | -0.154 |
Token | play |
Token ID | 711 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
ey | +0.019 |
babys | +0.019 |
tant | +0.019 |
priest | +0.018 |
hotel | +0.018 |
Neg logit contributions | |
there | -0.015 |
on | -0.015 |
in | -0.014 |
together | -0.014 |
with | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
icked | +0.044 |
es | +0.037 |
ppers | +0.033 |
iled | +0.031 |
owed | +0.029 |
Neg logit contributions | |
classroom | -0.082 |
museum | -0.082 |
playground | -0.081 |
hotel | -0.078 |
park | -0.078 |
Token | He |
Token ID | 679 |
Feature activation | -7.34e-03 |
Pos logit contributions | |
es | +0.017 |
together | +0.015 |
icked | +0.015 |
on | +0.015 |
there | +0.014 |
Neg logit contributions | |
hotel | -0.023 |
classroom | -0.022 |
ey | -0.022 |
city | -0.022 |
museum | -0.022 |
Token | happened |
Token ID | 3022 |
Feature activation | -4.64e-02 |
Pos logit contributions | |
hotel | +0.013 |
classroom | +0.012 |
principal | +0.012 |
city | +0.012 |
museum | +0.012 |
Neg logit contributions | |
es | -0.009 |
icked | -0.008 |
together | -0.008 |
on | -0.007 |
there | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -7.44e-03 |
Pos logit contributions | |
icked | +0.067 |
es | +0.060 |
ided | +0.044 |
pped | +0.044 |
apy | +0.044 |
Neg logit contributions | |
classroom | -0.098 |
principal | -0.095 |
talent | -0.094 |
city | -0.093 |
museum | -0.093 |
Token | Her |
Token ID | 2332 |
Feature activation | +7.80e-02 |
Pos logit contributions | |
es | +0.010 |
on | +0.009 |
together | +0.009 |
icked | +0.009 |
there | +0.009 |
Neg logit contributions | |
hotel | -0.017 |
ey | -0.017 |
city | -0.016 |
principal | -0.016 |
classroom | -0.016 |
Token | mom |
Token ID | 1995 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
museum | +0.167 |
orchestra | +0.167 |
hotel | +0.166 |
carn | +0.166 |
city | +0.164 |
Neg logit contributions | |
on | -0.116 |
there | -0.114 |
in | -0.100 |
es | -0.099 |
with | -0.095 |
Token | hugged |
Token ID | 41130 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
es | +0.039 |
icked | +0.036 |
on | +0.036 |
together | +0.035 |
there | +0.035 |
Neg logit contributions | |
hotel | -0.056 |
principal | -0.054 |
city | -0.054 |
ey | -0.054 |
classroom | -0.053 |
Token | her |
Token ID | 607 |
Feature activation | +4.10e-02 |
Pos logit contributions | |
es | +0.047 |
icked | +0.042 |
on | +0.040 |
together | +0.039 |
there | +0.039 |
Neg logit contributions | |
hotel | -0.080 |
ey | -0.078 |
classroom | -0.077 |
principal | -0.077 |
city | -0.077 |
Token | and |
Token ID | 290 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
hotel | +0.116 |
ey | +0.113 |
city | +0.112 |
classroom | +0.112 |
principal | +0.112 |
Neg logit contributions | |
es | -0.036 |
on | -0.030 |
icked | -0.030 |
together | -0.029 |
there | -0.028 |
Token | said |
Token ID | 531 |
Feature activation | -3.14e-02 |
Pos logit contributions | |
es | +0.032 |
on | +0.028 |
icked | +0.028 |
together | +0.027 |
there | +0.026 |
Neg logit contributions | |
hotel | -0.055 |
ey | -0.053 |
principal | -0.053 |
classroom | -0.053 |
city | -0.053 |
Token | , |
Token ID | 11 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
es | +0.035 |
against | +0.033 |
icked | +0.033 |
together | +0.029 |
perched | +0.028 |
Neg logit contributions | |
babys | -0.067 |
mirror | -0.064 |
hotel | -0.062 |
principal | -0.062 |
priest | -0.061 |
Token | " |
Token ID | 366 |
Feature activation | -2.71e-03 |
Pos logit contributions | |
es | +0.052 |
icked | +0.049 |
together | +0.043 |
there | +0.042 |
on | +0.042 |
Neg logit contributions | |
principal | -0.066 |
babys | -0.065 |
city | -0.063 |
hotel | -0.063 |
ey | -0.062 |
Token | Don |
Token ID | 3987 |
Feature activation | -9.93e-03 |
Pos logit contributions | |
es | +0.003 |
on | +0.003 |
together | +0.003 |
there | +0.003 |
icked | +0.003 |
Neg logit contributions | |
hotel | -0.007 |
ey | -0.006 |
city | -0.006 |
classroom | -0.006 |
principal | -0.006 |
Token | along |
Token ID | 1863 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
icked | +0.046 |
es | +0.046 |
there | +0.036 |
on | +0.035 |
together | +0.032 |
Neg logit contributions | |
hotel | -0.115 |
city | -0.112 |
classroom | -0.112 |
orchestra | -0.111 |
playground | -0.110 |
Token | with |
Token ID | 351 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
es | +0.019 |
icked | +0.019 |
on | +0.014 |
there | +0.014 |
together | +0.014 |
Neg logit contributions | |
hotel | -0.058 |
city | -0.056 |
classroom | -0.056 |
museum | -0.055 |
playground | -0.055 |
Token | the |
Token ID | 262 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
hotel | +0.030 |
ey | +0.029 |
principal | +0.029 |
classroom | +0.029 |
city | +0.029 |
Neg logit contributions | |
es | -0.014 |
on | -0.013 |
together | -0.013 |
there | -0.012 |
icked | -0.012 |
Token | music |
Token ID | 2647 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
ey | +0.196 |
principal | +0.191 |
hotel | +0.187 |
courtroom | +0.176 |
Ta | +0.175 |
Neg logit contributions | |
on | -0.288 |
in | -0.263 |
there | -0.262 |
together | -0.256 |
with | -0.242 |
Token | . |
Token ID | 13 |
Feature activation | -9.86e-04 |
Pos logit contributions | |
es | +0.014 |
icked | +0.013 |
on | +0.011 |
there | +0.011 |
together | +0.011 |
Neg logit contributions | |
hotel | -0.043 |
classroom | -0.042 |
city | -0.042 |
playground | -0.041 |
principal | -0.041 |
Token | Grand |
Token ID | 5675 |
Feature activation | +8.58e-02 |
Pos logit contributions | |
es | +0.001 |
on | +0.001 |
together | +0.001 |
icked | +0.001 |
there | +0.001 |
Neg logit contributions | |
hotel | -0.002 |
ey | -0.002 |
city | -0.002 |
classroom | -0.002 |
principal | -0.002 |
Token | ma |
Token ID | 2611 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
hotel | +0.144 |
ey | +0.142 |
principal | +0.136 |
city | +0.136 |
classroom | +0.136 |
Neg logit contributions | |
es | -0.174 |
together | -0.161 |
on | -0.158 |
there | -0.156 |
icked | -0.153 |
Token | kept |
Token ID | 4030 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
es | +0.025 |
icked | +0.023 |
on | +0.023 |
together | +0.022 |
there | +0.022 |
Neg logit contributions | |
hotel | -0.040 |
city | -0.039 |
classroom | -0.039 |
principal | -0.039 |
ey | -0.038 |
Token | watching |
Token ID | 4964 |
Feature activation | -2.51e-02 |
Pos logit contributions | |
es | +0.032 |
icked | +0.029 |
together | +0.028 |
on | +0.028 |
there | +0.027 |
Neg logit contributions | |
hotel | -0.043 |
ey | -0.041 |
classroom | -0.041 |
city | -0.041 |
principal | -0.041 |
Token | while |
Token ID | 981 |
Feature activation | +4.98e-03 |
Pos logit contributions | |
es | +0.027 |
icked | +0.024 |
on | +0.023 |
together | +0.022 |
there | +0.022 |
Neg logit contributions | |
hotel | -0.065 |
ey | -0.063 |
city | -0.063 |
classroom | -0.062 |
principal | -0.062 |
Token | she |
Token ID | 673 |
Feature activation | +3.02e-02 |
Pos logit contributions | |
hotel | +0.012 |
city | +0.012 |
classroom | +0.011 |
principal | +0.011 |
museum | +0.011 |
Neg logit contributions | |
es | -0.006 |
icked | -0.006 |
together | -0.005 |
on | -0.005 |
there | -0.005 |
Token | creative |
Token ID | 7325 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
hotel | +0.025 |
ey | +0.024 |
city | +0.024 |
classroom | +0.024 |
principal | +0.024 |
Neg logit contributions | |
es | -0.030 |
on | -0.028 |
together | -0.028 |
icked | -0.028 |
there | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -4.25e-04 |
Pos logit contributions | |
icked | +0.022 |
es | +0.020 |
together | +0.016 |
there | +0.016 |
on | +0.015 |
Neg logit contributions | |
hotel | -0.053 |
classroom | -0.051 |
principal | -0.051 |
museum | -0.051 |
city | -0.050 |
Token | |
Token ID | 220 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
es | +0.001 |
on | +0.001 |
icked | +0.001 |
together | +0.001 |
there | +0.001 |
Neg logit contributions | |
hotel | -0.001 |
ey | -0.001 |
classroom | -0.001 |
city | -0.001 |
principal | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +7.85e-03 |
Pos logit contributions | |
hotel | +0.048 |
ey | +0.047 |
classroom | +0.047 |
principal | +0.046 |
city | +0.046 |
Neg logit contributions | |
es | -0.028 |
on | -0.026 |
icked | -0.025 |
together | -0.025 |
there | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | +2.52e-03 |
Pos logit contributions | |
hotel | +0.017 |
ey | +0.017 |
classroom | +0.017 |
city | +0.016 |
principal | +0.016 |
Neg logit contributions | |
es | -0.011 |
icked | -0.010 |
on | -0.010 |
together | -0.010 |
there | -0.009 |
Token | M |
Token ID | 44 |
Feature activation | +7.88e-02 |
Pos logit contributions | |
hotel | +0.006 |
ey | +0.005 |
classroom | +0.005 |
city | +0.005 |
principal | +0.005 |
Neg logit contributions | |
es | -0.004 |
icked | -0.003 |
on | -0.003 |
together | -0.003 |
there | -0.003 |
Token | andy |
Token ID | 10757 |
Feature activation | -5.01e-03 |
Pos logit contributions | |
hotel | +0.134 |
city | +0.133 |
principal | +0.130 |
playground | +0.130 |
orchestra | +0.128 |
Neg logit contributions | |
es | -0.153 |
icked | -0.147 |
on | -0.141 |
together | -0.140 |
there | -0.138 |
Token | smiled |
Token ID | 13541 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
es | +0.007 |
on | +0.006 |
together | +0.006 |
icked | +0.006 |
there | +0.006 |
Neg logit contributions | |
hotel | -0.011 |
ey | -0.011 |
city | -0.011 |
classroom | -0.011 |
principal | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
es | +0.031 |
icked | +0.028 |
on | +0.023 |
there | +0.023 |
together | +0.022 |
Neg logit contributions | |
hotel | -0.067 |
classroom | -0.066 |
ey | -0.066 |
city | -0.066 |
former | -0.065 |
Token | hugged |
Token ID | 41130 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
es | +0.022 |
icked | +0.020 |
on | +0.019 |
there | +0.019 |
together | +0.019 |
Neg logit contributions | |
hotel | -0.028 |
classroom | -0.027 |
ey | -0.027 |
playground | -0.027 |
city | -0.027 |
Token | her |
Token ID | 607 |
Feature activation | +8.35e-02 |
Pos logit contributions | |
es | +0.027 |
icked | +0.024 |
on | +0.024 |
together | +0.023 |
there | +0.023 |
Neg logit contributions | |
hotel | -0.054 |
ey | -0.053 |
classroom | -0.052 |
city | -0.052 |
principal | -0.052 |
Token | and |
Token ID | 290 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
es | +0.033 |
icked | +0.029 |
on | +0.029 |
together | +0.028 |
there | +0.027 |
Neg logit contributions | |
hotel | -0.070 |
ey | -0.068 |
city | -0.068 |
classroom | -0.067 |
principal | -0.067 |
Token | Julia |
Token ID | 22300 |
Feature activation | -9.72e-03 |
Pos logit contributions | |
es | +0.012 |
icked | +0.010 |
on | +0.010 |
together | +0.010 |
there | +0.010 |
Neg logit contributions | |
hotel | -0.026 |
ey | -0.025 |
classroom | -0.025 |
city | -0.025 |
principal | -0.025 |
Token | loved |
Token ID | 6151 |
Feature activation | -7.22e-03 |
Pos logit contributions | |
es | +0.014 |
icked | +0.013 |
on | +0.013 |
together | +0.013 |
there | +0.013 |
Neg logit contributions | |
hotel | -0.021 |
city | -0.020 |
ey | -0.020 |
classroom | -0.020 |
principal | -0.020 |
Token | watching |
Token ID | 4964 |
Feature activation | -2.33e-02 |
Pos logit contributions | |
es | +0.009 |
icked | +0.008 |
on | +0.007 |
together | +0.007 |
there | +0.007 |
Neg logit contributions | |
hotel | -0.018 |
city | -0.018 |
classroom | -0.017 |
ey | -0.017 |
museum | -0.017 |
Token | it |
Token ID | 340 |
Feature activation | -2.28e-04 |
Pos logit contributions | |
es | +0.027 |
icked | +0.024 |
on | +0.023 |
together | +0.022 |
there | +0.022 |
Neg logit contributions | |
hotel | -0.059 |
ey | -0.057 |
city | -0.057 |
classroom | -0.056 |
principal | -0.056 |
Token | each |
Token ID | 1123 |
Feature activation | +5.40e-02 |
Pos logit contributions | |
es | +0.000 |
icked | +0.000 |
there | +0.000 |
on | +0.000 |
together | +0.000 |
Neg logit contributions | |
hotel | -0.000 |
city | -0.000 |
classroom | -0.000 |
playground | -0.000 |
former | -0.000 |
Token | night |
Token ID | 1755 |
Feature activation | -8.35e-03 |
Pos logit contributions | |
hotel | +0.071 |
ey | +0.070 |
principal | +0.067 |
classroom | +0.067 |
city | +0.067 |
Neg logit contributions | |
es | -0.127 |
on | -0.121 |
together | -0.119 |
there | -0.118 |
icked | -0.118 |
Token | . |
Token ID | 13 |
Feature activation | +7.77e-03 |
Pos logit contributions | |
es | +0.008 |
icked | +0.007 |
on | +0.006 |
together | +0.006 |
there | +0.006 |
Neg logit contributions | |
hotel | -0.023 |
city | -0.022 |
classroom | -0.022 |
ey | -0.022 |
playground | -0.022 |
Token | No |
Token ID | 1400 |
Feature activation | +4.77e-02 |
Pos logit contributions | |
hotel | +0.021 |
ey | +0.020 |
principal | +0.020 |
city | +0.019 |
orchestra | +0.019 |
Neg logit contributions | |
es | -0.008 |
on | -0.007 |
together | -0.007 |
there | -0.007 |
icked | -0.007 |
Token | matter |
Token ID | 2300 |
Feature activation | -2.30e-02 |
Pos logit contributions | |
ey | +0.083 |
hotel | +0.081 |
principal | +0.080 |
city | +0.078 |
classroom | +0.078 |
Neg logit contributions | |
es | -0.091 |
on | -0.089 |
together | -0.087 |
there | -0.086 |
instead | -0.082 |
Token | how |
Token ID | 703 |
Feature activation | +2.83e-03 |
Pos logit contributions | |
es | +0.027 |
icked | +0.026 |
on | +0.022 |
against | +0.021 |
together | +0.020 |
Neg logit contributions | |
hotel | -0.055 |
city | -0.054 |
classroom | -0.053 |
museum | -0.053 |
mall | -0.053 |
Token | play |
Token ID | 711 |
Feature activation | -3.02e-02 |
Pos logit contributions | |
es | +0.005 |
icked | +0.005 |
on | +0.005 |
together | +0.005 |
there | +0.005 |
Neg logit contributions | |
hotel | -0.005 |
classroom | -0.005 |
city | -0.005 |
principal | -0.005 |
ey | -0.005 |
Token | somewhere |
Token ID | 7382 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
icked | +0.027 |
es | +0.025 |
on | +0.017 |
there | +0.016 |
able | +0.015 |
Neg logit contributions | |
hotel | -0.083 |
classroom | -0.083 |
museum | -0.081 |
playground | -0.081 |
city | -0.080 |
Token | else |
Token ID | 2073 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
es | +0.024 |
icked | +0.023 |
on | +0.019 |
together | +0.019 |
there | +0.018 |
Neg logit contributions | |
hotel | -0.027 |
classroom | -0.025 |
museum | -0.025 |
city | -0.025 |
principal | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -2.17e-03 |
Pos logit contributions | |
icked | +0.015 |
es | +0.012 |
pped | +0.010 |
on | +0.010 |
against | +0.010 |
Neg logit contributions | |
hotel | -0.027 |
classroom | -0.027 |
museum | -0.026 |
city | -0.026 |
principal | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | -8.57e-04 |
Pos logit contributions | |
es | +0.003 |
icked | +0.003 |
on | +0.003 |
there | +0.003 |
together | +0.003 |
Neg logit contributions | |
hotel | -0.005 |
classroom | -0.004 |
city | -0.004 |
principal | -0.004 |
museum | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -7.72e-03 |
Pos logit contributions | |
es | +0.001 |
on | +0.001 |
together | +0.001 |
there | +0.001 |
icked | +0.001 |
Neg logit contributions | |
hotel | -0.002 |
ey | -0.002 |
city | -0.002 |
classroom | -0.002 |
principal | -0.002 |
Token | L |
Token ID | 43 |
Feature activation | +7.04e-02 |
Pos logit contributions | |
es | +0.013 |
on | +0.012 |
together | +0.012 |
there | +0.012 |
icked | +0.012 |
Neg logit contributions | |
hotel | -0.015 |
ey | -0.015 |
city | -0.015 |
classroom | -0.015 |
principal | -0.015 |
Token | ily |
Token ID | 813 |
Feature activation | -1.00e-02 |
Pos logit contributions | |
hotel | +0.158 |
ey | +0.154 |
city | +0.152 |
classroom | +0.151 |
principal | +0.150 |
Neg logit contributions | |
es | -0.104 |
on | -0.094 |
icked | -0.093 |
together | -0.092 |
there | -0.091 |
Token | is |
Token ID | 318 |
Feature activation | +3.22e-03 |
Pos logit contributions | |
es | +0.013 |
on | +0.012 |
together | +0.011 |
there | +0.011 |
icked | +0.011 |
Neg logit contributions | |
hotel | -0.025 |
ey | -0.024 |
city | -0.024 |
classroom | -0.024 |
museum | -0.023 |
Token | angry |
Token ID | 7954 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
hotel | +0.005 |
ey | +0.005 |
classroom | +0.005 |
principal | +0.005 |
city | +0.005 |
Neg logit contributions | |
es | -0.007 |
icked | -0.006 |
on | -0.006 |
together | -0.006 |
there | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -8.15e-04 |
Pos logit contributions | |
icked | +0.049 |
es | +0.045 |
bled | +0.039 |
fed | +0.038 |
able | +0.038 |
Neg logit contributions | |
classroom | -0.061 |
principal | -0.060 |
museum | -0.058 |
mirror | -0.057 |
talent | -0.056 |
Token | her |
Token ID | 607 |
Feature activation | +8.39e-02 |
Pos logit contributions | |
es | +0.021 |
icked | +0.018 |
on | +0.018 |
together | +0.017 |
there | +0.017 |
Neg logit contributions | |
hotel | -0.041 |
principal | -0.040 |
ey | -0.040 |
classroom | -0.040 |
city | -0.039 |
Token | mom |
Token ID | 1995 |
Feature activation | -3.07e-02 |
Pos logit contributions | |
ey | +0.185 |
museum | +0.181 |
carn | +0.179 |
hotel | +0.179 |
city | +0.178 |
Neg logit contributions | |
es | -0.129 |
there | -0.122 |
on | -0.122 |
together | -0.115 |
icked | -0.115 |
Token | said |
Token ID | 531 |
Feature activation | -4.37e-02 |
Pos logit contributions | |
es | +0.047 |
icked | +0.043 |
on | +0.043 |
there | +0.041 |
together | +0.040 |
Neg logit contributions | |
principal | -0.059 |
hotel | -0.058 |
babys | -0.058 |
priest | -0.058 |
classroom | -0.057 |
Token | . |
Token ID | 13 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
icked | +0.051 |
es | +0.051 |
there | +0.041 |
on | +0.039 |
pped | +0.038 |
Neg logit contributions | |
babys | -0.092 |
principal | -0.088 |
priest | -0.087 |
doorway | -0.087 |
salon | -0.085 |
Token | " |
Token ID | 366 |
Feature activation | -4.02e-03 |
Pos logit contributions | |
es | +0.019 |
on | +0.017 |
icked | +0.017 |
there | +0.015 |
against | +0.015 |
Neg logit contributions | |
classroom | -0.027 |
principal | -0.027 |
hotel | -0.026 |
babys | -0.026 |
playground | -0.026 |
Token | We |
Token ID | 1135 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
es | +0.005 |
together | +0.004 |
on | +0.004 |
there | +0.004 |
instead | +0.004 |
Neg logit contributions | |
hotel | -0.010 |
ey | -0.010 |
city | -0.010 |
classroom | -0.010 |
principal | -0.010 |
Token | can |
Token ID | 460 |
Feature activation | -6.96e-03 |
Pos logit contributions | |
hotel | +0.028 |
ey | +0.028 |
principal | +0.028 |
classroom | +0.027 |
city | +0.027 |
Neg logit contributions | |
es | -0.015 |
there | -0.015 |
on | -0.015 |
together | -0.014 |
icked | -0.014 |
Token | get |
Token ID | 651 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
es | +0.012 |
on | +0.011 |
together | +0.011 |
icked | +0.011 |
there | +0.011 |
Neg logit contributions | |
hotel | -0.014 |
ey | -0.013 |
city | -0.013 |
classroom | -0.013 |
principal | -0.013 |
Token | you |
Token ID | 345 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
es | +0.020 |
icked | +0.016 |
on | +0.015 |
together | +0.015 |
there | +0.014 |
Neg logit contributions | |
hotel | -0.058 |
ey | -0.056 |
principal | -0.056 |
classroom | -0.056 |
city | -0.056 |
Token | a |
Token ID | 257 |
Feature activation | +8.64e-02 |
Pos logit contributions | |
hotel | +0.040 |
city | +0.039 |
classroom | +0.039 |
ey | +0.039 |
playground | +0.038 |
Neg logit contributions | |
es | -0.023 |
icked | -0.020 |
on | -0.019 |
together | -0.018 |
there | -0.018 |
Token | new |
Token ID | 649 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
ey | +0.159 |
principal | +0.152 |
hotel | +0.150 |
city | +0.150 |
orchestra | +0.148 |
Neg logit contributions | |
on | -0.172 |
there | -0.160 |
together | -0.158 |
in | -0.153 |
with | -0.152 |
Token | the |
Token ID | 262 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
hotel | +0.044 |
ey | +0.043 |
principal | +0.042 |
city | +0.042 |
classroom | +0.042 |
Neg logit contributions | |
es | -0.016 |
on | -0.015 |
there | -0.014 |
together | -0.014 |
icked | -0.013 |
Token | hole |
Token ID | 7604 |
Feature activation | -6.06e-02 |
Pos logit contributions | |
orchestra | +0.190 |
priest | +0.188 |
pastor | +0.184 |
salon | +0.178 |
actresses | +0.177 |
Neg logit contributions | |
on | -0.247 |
in | -0.216 |
there | -0.213 |
down | -0.195 |
with | -0.194 |
Token | . |
Token ID | 13 |
Feature activation | -2.67e-04 |
Pos logit contributions | |
es | +0.060 |
icked | +0.057 |
there | +0.043 |
on | +0.042 |
together | +0.041 |
Neg logit contributions | |
hotel | -0.148 |
city | -0.142 |
doorway | -0.142 |
classroom | -0.141 |
playground | -0.141 |
Token | He |
Token ID | 679 |
Feature activation | -7.78e-03 |
Pos logit contributions | |
es | +0.000 |
on | +0.000 |
icked | +0.000 |
together | +0.000 |
there | +0.000 |
Neg logit contributions | |
hotel | -0.001 |
ey | -0.001 |
principal | -0.001 |
classroom | -0.001 |
city | -0.001 |
Token | said |
Token ID | 531 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
es | +0.010 |
on | +0.009 |
icked | +0.009 |
together | +0.009 |
there | +0.009 |
Neg logit contributions | |
hotel | -0.018 |
ey | -0.018 |
city | -0.018 |
classroom | -0.018 |
principal | -0.018 |
Token | no |
Token ID | 645 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
es | +0.021 |
icked | +0.018 |
together | +0.017 |
on | +0.016 |
against | +0.016 |
Neg logit contributions | |
hotel | -0.047 |
principal | -0.047 |
priest | -0.046 |
museum | -0.045 |
city | -0.045 |
Token | one |
Token ID | 530 |
Feature activation | +1.28e-02 |
Pos logit contributions | |
hotel | +0.060 |
ey | +0.059 |
city | +0.058 |
principal | +0.058 |
classroom | +0.058 |
Neg logit contributions | |
es | -0.014 |
on | -0.012 |
together | -0.011 |
there | -0.011 |
icked | -0.010 |
Token | should |
Token ID | 815 |
Feature activation | -1.83e-03 |
Pos logit contributions | |
hotel | +0.029 |
city | +0.028 |
museum | +0.028 |
ey | +0.028 |
principal | +0.028 |
Neg logit contributions | |
es | -0.017 |
icked | -0.016 |
together | -0.015 |
on | -0.015 |
there | -0.015 |
Token | ever |
Token ID | 1683 |
Feature activation | +4.07e-03 |
Pos logit contributions | |
es | +0.003 |
on | +0.003 |
icked | +0.003 |
together | +0.003 |
there | +0.003 |
Neg logit contributions | |
hotel | -0.004 |
ey | -0.003 |
city | -0.003 |
classroom | -0.003 |
principal | -0.003 |
Token | touch |
Token ID | 3638 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
hotel | +0.007 |
ey | +0.007 |
city | +0.007 |
classroom | +0.007 |
principal | +0.007 |
Neg logit contributions | |
es | -0.008 |
icked | -0.007 |
on | -0.007 |
together | -0.007 |
there | -0.007 |
Token | it |
Token ID | 340 |
Feature activation | -4.64e-03 |
Pos logit contributions | |
es | +0.017 |
icked | +0.016 |
together | +0.015 |
on | +0.013 |
there | +0.012 |
Neg logit contributions | |
hotel | -0.051 |
city | -0.050 |
ey | -0.050 |
museum | -0.050 |
classroom | -0.049 |
Token | happy |
Token ID | 3772 |
Feature activation | -9.90e-03 |
Pos logit contributions | |
es | +0.025 |
on | +0.024 |
together | +0.024 |
there | +0.023 |
icked | +0.023 |
Neg logit contributions | |
hotel | -0.024 |
ey | -0.023 |
city | -0.023 |
principal | -0.023 |
classroom | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -3.88e-03 |
Pos logit contributions | |
es | +0.011 |
icked | +0.011 |
pped | +0.008 |
on | +0.008 |
there | +0.008 |
Neg logit contributions | |
hotel | -0.025 |
classroom | -0.024 |
museum | -0.024 |
principal | -0.024 |
city | -0.024 |
Token | They |
Token ID | 1119 |
Feature activation | -2.31e-03 |
Pos logit contributions | |
es | +0.005 |
on | +0.004 |
icked | +0.004 |
together | +0.004 |
there | +0.004 |
Neg logit contributions | |
hotel | -0.009 |
ey | -0.009 |
city | -0.009 |
classroom | -0.009 |
principal | -0.009 |
Token | tell |
Token ID | 1560 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
on | +0.004 |
there | +0.003 |
in | +0.003 |
together | +0.003 |
es | +0.003 |
Neg logit contributions | |
ey | -0.005 |
doorway | -0.005 |
former | -0.005 |
hallway | -0.005 |
hotel | -0.005 |
Token | them |
Token ID | 606 |
Feature activation | -7.10e-03 |
Pos logit contributions | |
es | +0.020 |
on | +0.018 |
together | +0.017 |
there | +0.017 |
icked | +0.017 |
Neg logit contributions | |
ey | -0.046 |
hotel | -0.046 |
classroom | -0.045 |
city | -0.045 |
principal | -0.044 |
Token | that |
Token ID | 326 |
Feature activation | -4.24e-03 |
Pos logit contributions | |
es | +0.007 |
icked | +0.007 |
on | +0.006 |
together | +0.006 |
there | +0.006 |
Neg logit contributions | |
hotel | -0.018 |
city | -0.018 |
museum | -0.018 |
classroom | -0.018 |
principal | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
es | +0.003 |
icked | +0.003 |
together | +0.003 |
on | +0.003 |
there | +0.002 |
Neg logit contributions | |
hotel | -0.012 |
city | -0.012 |
classroom | -0.012 |
principal | -0.012 |
museum | -0.012 |
Token | envelope |
Token ID | 22878 |
Feature activation | -4.54e-02 |
Pos logit contributions | |
opl | +0.247 |
orchestra | +0.245 |
hotel | +0.244 |
apartment | +0.241 |
igsaw | +0.241 |
Neg logit contributions | |
there | -0.291 |
on | -0.286 |
in | -0.258 |
with | -0.247 |
it | -0.243 |
Token | and |
Token ID | 290 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
es | +0.079 |
icked | +0.072 |
together | +0.072 |
on | +0.072 |
there | +0.070 |
Neg logit contributions | |
hotel | -0.086 |
ey | -0.083 |
classroom | -0.082 |
city | -0.082 |
principal | -0.082 |
Token | the |
Token ID | 262 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
hotel | +0.047 |
ey | +0.046 |
city | +0.045 |
classroom | +0.045 |
principal | +0.045 |
Neg logit contributions | |
es | -0.017 |
on | -0.016 |
together | -0.015 |
there | -0.015 |
icked | -0.015 |
Token | paper |
Token ID | 3348 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
hotel | +0.201 |
igsaw | +0.195 |
opl | +0.192 |
ey | +0.191 |
orchestra | +0.188 |
Neg logit contributions | |
on | -0.277 |
there | -0.266 |
in | -0.251 |
together | -0.236 |
with | -0.235 |
Token | " |
Token ID | 366 |
Feature activation | +4.82e-03 |
Pos logit contributions | |
es | +0.056 |
icked | +0.052 |
together | +0.047 |
there | +0.046 |
led | +0.044 |
Neg logit contributions | |
principal | -0.051 |
babys | -0.051 |
zoo | -0.051 |
bus | -0.049 |
mirror | -0.049 |
Token | You |
Token ID | 1639 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
hotel | +0.010 |
ey | +0.010 |
city | +0.010 |
classroom | +0.010 |
principal | +0.010 |
Neg logit contributions | |
es | -0.007 |
on | -0.007 |
icked | -0.007 |
together | -0.007 |
there | -0.007 |
Token | are |
Token ID | 389 |
Feature activation | +1.41e-03 |
Pos logit contributions | |
hotel | +0.039 |
ey | +0.038 |
city | +0.037 |
principal | +0.037 |
classroom | +0.037 |
Neg logit contributions | |
es | -0.036 |
on | -0.034 |
icked | -0.034 |
there | -0.034 |
together | -0.033 |
Token | welcome |
Token ID | 7062 |
Feature activation | -7.68e-03 |
Pos logit contributions | |
hotel | +0.002 |
ey | +0.002 |
classroom | +0.002 |
principal | +0.002 |
museum | +0.002 |
Neg logit contributions | |
es | -0.003 |
icked | -0.003 |
together | -0.003 |
on | -0.002 |
there | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -9.64e-03 |
Pos logit contributions | |
es | +0.009 |
icked | +0.009 |
together | +0.008 |
ged | +0.007 |
against | +0.007 |
Neg logit contributions | |
hotel | -0.018 |
babys | -0.018 |
mirror | -0.017 |
audience | -0.017 |
zoo | -0.017 |
Token | You |
Token ID | 921 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
es | +0.015 |
on | +0.014 |
together | +0.014 |
icked | +0.013 |
there | +0.013 |
Neg logit contributions | |
hotel | -0.020 |
ey | -0.020 |
classroom | -0.019 |
museum | -0.019 |
city | -0.019 |
Token | are |
Token ID | 389 |
Feature activation | +2.81e-03 |
Pos logit contributions | |
hotel | +0.023 |
ey | +0.023 |
city | +0.022 |
classroom | +0.022 |
principal | +0.022 |
Neg logit contributions | |
es | -0.021 |
on | -0.019 |
icked | -0.019 |
together | -0.019 |
there | -0.019 |
Token | my |
Token ID | 616 |
Feature activation | +9.31e-02 |
Pos logit contributions | |
hotel | +0.006 |
classroom | +0.006 |
ey | +0.006 |
principal | +0.006 |
city | +0.006 |
Neg logit contributions | |
es | -0.005 |
icked | -0.004 |
together | -0.004 |
on | -0.004 |
there | -0.004 |
Token | friend |
Token ID | 1545 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
ey | +0.189 |
hallway | +0.170 |
doorway | +0.166 |
courtroom | +0.165 |
Ta | +0.165 |
Neg logit contributions | |
on | -0.171 |
together | -0.169 |
there | -0.165 |
in | -0.161 |
at | -0.158 |
Token | , |
Token ID | 11 |
Feature activation | +5.27e-04 |
Pos logit contributions | |
icked | +0.019 |
ged | +0.017 |
es | +0.017 |
bed | +0.016 |
pped | +0.015 |
Neg logit contributions | |
hotel | -0.033 |
audience | -0.033 |
city | -0.033 |
zoo | -0.032 |
museum | -0.032 |
Token | too |
Token ID | 1165 |
Feature activation | -3.13e-03 |
Pos logit contributions | |
hotel | +0.001 |
ey | +0.001 |
city | +0.001 |
classroom | +0.001 |
principal | +0.001 |
Neg logit contributions | |
es | -0.001 |
on | -0.001 |
icked | -0.001 |
together | -0.001 |
there | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
es | +0.045 |
icked | +0.039 |
together | +0.036 |
on | +0.036 |
there | +0.035 |
Neg logit contributions | |
hotel | -0.104 |
ey | -0.101 |
classroom | -0.101 |
city | -0.100 |
principal | -0.100 |
Token | went |
Token ID | 1816 |
Feature activation | -3.64e-02 |
Pos logit contributions | |
es | +0.051 |
icked | +0.049 |
iled | +0.040 |
together | +0.040 |
on | +0.039 |
Neg logit contributions | |
hotel | -0.054 |
classroom | -0.052 |
mirror | -0.052 |
airport | -0.051 |
beach | -0.051 |
Token | back |
Token ID | 736 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
es | +0.039 |
icked | +0.034 |
together | +0.030 |
on | +0.029 |
there | +0.029 |
Neg logit contributions | |
hotel | -0.098 |
classroom | -0.094 |
principal | -0.093 |
city | -0.093 |
playground | -0.092 |
Token | to |
Token ID | 284 |
Feature activation | -7.24e-03 |
Pos logit contributions | |
es | +0.029 |
icked | +0.027 |
together | +0.020 |
on | +0.020 |
instead | +0.020 |
Neg logit contributions | |
hotel | -0.060 |
classroom | -0.058 |
city | -0.057 |
playground | -0.056 |
museum | -0.056 |
Token | the |
Token ID | 262 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
es | +0.015 |
icked | +0.014 |
together | +0.013 |
on | +0.013 |
there | +0.013 |
Neg logit contributions | |
hotel | -0.012 |
classroom | -0.012 |
city | -0.011 |
playground | -0.011 |
principal | -0.011 |
Token | porch |
Token ID | 33179 |
Feature activation | -6.49e-02 |
Pos logit contributions | |
orchestra | +0.161 |
actor | +0.160 |
igsaw | +0.159 |
ey | +0.158 |
tant | +0.155 |
Neg logit contributions | |
on | -0.273 |
there | -0.271 |
in | -0.256 |
with | -0.254 |
at | -0.252 |
Token | to |
Token ID | 284 |
Feature activation | +2.05e-03 |
Pos logit contributions | |
es | +0.058 |
icked | +0.051 |
on | +0.045 |
together | +0.043 |
there | +0.042 |
Neg logit contributions | |
hotel | -0.182 |
classroom | -0.178 |
city | -0.177 |
principal | -0.176 |
playground | -0.175 |
Token | play |
Token ID | 711 |
Feature activation | -3.95e-02 |
Pos logit contributions | |
hotel | +0.003 |
ey | +0.003 |
city | +0.003 |
principal | +0.003 |
classroom | +0.003 |
Neg logit contributions | |
es | -0.004 |
on | -0.004 |
there | -0.004 |
together | -0.004 |
icked | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
icked | +0.053 |
es | +0.042 |
iled | +0.038 |
ppers | +0.037 |
neither | +0.035 |
Neg logit contributions | |
hotel | -0.095 |
classroom | -0.094 |
playground | -0.092 |
theater | -0.091 |
hall | -0.090 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
es | +0.030 |
on | +0.027 |
together | +0.026 |
icked | +0.026 |
there | +0.026 |
Neg logit contributions | |
hotel | -0.052 |
ey | -0.051 |
city | -0.050 |
principal | -0.050 |
classroom | -0.050 |
Token | , |
Token ID | 11 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
icked | +0.037 |
es | +0.036 |
together | +0.030 |
against | +0.027 |
iled | +0.027 |
Neg logit contributions | |
hotel | -0.081 |
classroom | -0.077 |
city | -0.077 |
airport | -0.075 |
babys | -0.075 |
Token | !" |
Token ID | 2474 |
Feature activation | -7.99e-03 |
Pos logit contributions | |
hotel | +0.002 |
city | +0.002 |
classroom | +0.002 |
ey | +0.002 |
playground | +0.002 |
Neg logit contributions | |
es | -0.001 |
icked | -0.001 |
together | -0.001 |
on | -0.001 |
there | -0.001 |
Token | Daisy |
Token ID | 40355 |
Feature activation | -3.24e-02 |
Pos logit contributions | |
es | +0.012 |
icked | +0.012 |
on | +0.011 |
there | +0.010 |
together | +0.010 |
Neg logit contributions | |
ey | -0.016 |
doorway | -0.016 |
hotel | -0.016 |
playground | -0.016 |
classroom | -0.016 |
Token | was |
Token ID | 373 |
Feature activation | -8.98e-04 |
Pos logit contributions | |
es | +0.049 |
icked | +0.045 |
on | +0.045 |
together | +0.045 |
there | +0.044 |
Neg logit contributions | |
hotel | -0.067 |
principal | -0.064 |
ey | -0.064 |
city | -0.064 |
classroom | -0.063 |
Token | scared |
Token ID | 12008 |
Feature activation | -4.09e-02 |
Pos logit contributions | |
es | +0.002 |
icked | +0.001 |
together | +0.001 |
on | +0.001 |
there | +0.001 |
Neg logit contributions | |
hotel | -0.002 |
ey | -0.002 |
principal | -0.002 |
classroom | -0.002 |
city | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
es | +0.059 |
icked | +0.057 |
together | +0.042 |
owed | +0.038 |
bed | +0.037 |
Neg logit contributions | |
hotel | -0.090 |
mirror | -0.088 |
city | -0.088 |
ey | -0.088 |
classroom | -0.087 |
Token | started |
Token ID | 2067 |
Feature activation | -4.18e-02 |
Pos logit contributions | |
es | +0.026 |
icked | +0.023 |
together | +0.023 |
on | +0.022 |
there | +0.021 |
Neg logit contributions | |
hotel | -0.036 |
principal | -0.034 |
city | -0.034 |
classroom | -0.034 |
ey | -0.034 |
Token | to |
Token ID | 284 |
Feature activation | -8.98e-03 |
Pos logit contributions | |
es | +0.039 |
icked | +0.033 |
together | +0.031 |
on | +0.031 |
there | +0.030 |
Neg logit contributions | |
hotel | -0.115 |
ey | -0.112 |
classroom | -0.112 |
principal | -0.112 |
city | -0.111 |
Token | cry |
Token ID | 3960 |
Feature activation | -5.16e-02 |
Pos logit contributions | |
es | +0.019 |
on | +0.019 |
there | +0.019 |
together | +0.018 |
icked | +0.018 |
Neg logit contributions | |
hotel | -0.013 |
ey | -0.013 |
principal | -0.012 |
city | -0.012 |
classroom | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
icked | +0.066 |
es | +0.063 |
pped | +0.053 |
there | +0.052 |
on | +0.045 |
Neg logit contributions | |
hotel | -0.113 |
babys | -0.111 |
former | -0.109 |
classroom | -0.108 |
city | -0.106 |
Token | |
Token ID | 220 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
es | +0.023 |
icked | +0.021 |
on | +0.021 |
together | +0.020 |
there | +0.019 |
Neg logit contributions | |
classroom | -0.027 |
hotel | -0.027 |
city | -0.027 |
playground | -0.026 |
ey | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | -4.61e-03 |
Pos logit contributions | |
classroom | +0.022 |
principal | +0.022 |
hotel | +0.022 |
ey | +0.022 |
city | +0.022 |
Neg logit contributions | |
es | -0.018 |
icked | -0.017 |
on | -0.017 |
together | -0.016 |
there | -0.015 |
Token | filled |
Token ID | 5901 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
es | +0.014 |
icked | +0.012 |
together | +0.011 |
on | +0.011 |
there | +0.010 |
Neg logit contributions | |
hotel | -0.018 |
classroom | -0.018 |
doorway | -0.018 |
ey | -0.018 |
playground | -0.018 |
Token | with |
Token ID | 351 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
icked | +0.014 |
es | +0.013 |
on | +0.007 |
there | +0.007 |
together | +0.006 |
Neg logit contributions | |
hotel | -0.047 |
museum | -0.046 |
beach | -0.046 |
city | -0.046 |
classroom | -0.046 |
Token | beautiful |
Token ID | 4950 |
Feature activation | +4.64e-02 |
Pos logit contributions | |
principal | +0.047 |
hotel | +0.046 |
ey | +0.044 |
salon | +0.044 |
classroom | +0.044 |
Neg logit contributions | |
on | -0.050 |
together | -0.049 |
es | -0.049 |
there | -0.049 |
in | -0.047 |
Token | butterflies |
Token ID | 44571 |
Feature activation | -5.99e-02 |
Pos logit contributions | |
hotel | +0.070 |
principal | +0.069 |
ey | +0.065 |
classroom | +0.065 |
city | +0.065 |
Neg logit contributions | |
es | -0.102 |
on | -0.097 |
together | -0.095 |
there | -0.094 |
icked | -0.089 |
Token | and |
Token ID | 290 |
Feature activation | +5.99e-04 |
Pos logit contributions | |
icked | +0.063 |
es | +0.058 |
there | +0.044 |
together | +0.043 |
instead | +0.041 |
Neg logit contributions | |
playground | -0.147 |
hotel | -0.146 |
beach | -0.145 |
city | -0.145 |
classroom | -0.142 |
Token | birds |
Token ID | 10087 |
Feature activation | -5.63e-02 |
Pos logit contributions | |
hotel | +0.001 |
principal | +0.001 |
ey | +0.001 |
classroom | +0.001 |
city | +0.001 |
Neg logit contributions | |
es | -0.001 |
on | -0.001 |
together | -0.001 |
there | -0.001 |
icked | -0.001 |
Token | ! |
Token ID | 0 |
Feature activation | -7.86e-03 |
Pos logit contributions | |
icked | +0.060 |
es | +0.060 |
there | +0.049 |
together | +0.046 |
instead | +0.046 |
Neg logit contributions | |
hotel | -0.130 |
playground | -0.129 |
city | -0.129 |
parade | -0.127 |
beach | -0.127 |
Token | Joey |
Token ID | 26154 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
es | +0.012 |
on | +0.011 |
icked | +0.011 |
together | +0.010 |
there | +0.010 |
Neg logit contributions | |
hotel | -0.017 |
ey | -0.017 |
city | -0.017 |
museum | -0.017 |
classroom | -0.016 |
Token | was |
Token ID | 373 |
Feature activation | -2.45e-04 |
Pos logit contributions | |
es | +0.019 |
icked | +0.018 |
on | +0.017 |
together | +0.017 |
there | +0.017 |
Neg logit contributions | |
hotel | -0.029 |
ey | -0.028 |
city | -0.028 |
classroom | -0.028 |
playground | -0.028 |
Token | so |
Token ID | 523 |
Feature activation | +7.28e-03 |
Pos logit contributions | |
es | +0.000 |
icked | +0.000 |
on | +0.000 |
together | +0.000 |
there | +0.000 |
Neg logit contributions | |
hotel | -0.001 |
ey | -0.001 |
classroom | -0.001 |
city | -0.001 |
principal | -0.001 |
Token | happy |
Token ID | 3772 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
hotel | +0.013 |
ey | +0.013 |
city | +0.012 |
classroom | +0.012 |
principal | +0.012 |
Neg logit contributions | |
es | -0.014 |
on | -0.013 |
icked | -0.013 |
together | -0.013 |
there | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
es | +0.015 |
icked | +0.014 |
on | +0.013 |
there | +0.012 |
together | +0.012 |
Neg logit contributions | |
hotel | -0.020 |
city | -0.020 |
classroom | -0.020 |
playground | -0.019 |
museum | -0.019 |
Token | Tim |
Token ID | 14967 |
Feature activation | +4.69e-03 |
Pos logit contributions | |
es | +0.026 |
icked | +0.025 |
on | +0.023 |
there | +0.022 |
together | +0.021 |
Neg logit contributions | |
playground | -0.026 |
city | -0.026 |
classroom | -0.025 |
hotel | -0.025 |
museum | -0.025 |
Token | my |
Token ID | 1820 |
Feature activation | -2.30e-02 |
Pos logit contributions | |
hotel | +0.008 |
ey | +0.008 |
classroom | +0.008 |
city | +0.008 |
principal | +0.007 |
Neg logit contributions | |
es | -0.009 |
on | -0.009 |
icked | -0.009 |
together | -0.009 |
there | -0.008 |
Token | obedient |
Token ID | 41652 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
es | +0.032 |
on | +0.029 |
icked | +0.029 |
together | +0.028 |
there | +0.028 |
Neg logit contributions | |
hotel | -0.052 |
classroom | -0.050 |
ey | -0.050 |
city | -0.050 |
principal | -0.050 |
Token | ly |
Token ID | 306 |
Feature activation | -4.93e-02 |
Pos logit contributions | |
hotel | +0.035 |
ey | +0.035 |
city | +0.034 |
principal | +0.033 |
classroom | +0.033 |
Neg logit contributions | |
es | -0.027 |
on | -0.025 |
there | -0.025 |
together | -0.025 |
icked | -0.024 |
Token | went |
Token ID | 1816 |
Feature activation | -3.79e-02 |
Pos logit contributions | |
es | +0.075 |
icked | +0.069 |
on | +0.069 |
there | +0.065 |
together | +0.064 |
Neg logit contributions | |
classroom | -0.097 |
hotel | -0.096 |
playground | -0.095 |
babys | -0.095 |
principal | -0.095 |
Token | to |
Token ID | 284 |
Feature activation | -7.77e-03 |
Pos logit contributions | |
on | +0.026 |
es | +0.026 |
together | +0.023 |
there | +0.022 |
in | +0.018 |
Neg logit contributions | |
ey | -0.113 |
hotel | -0.113 |
city | -0.110 |
former | -0.109 |
museum | -0.108 |
Token | his |
Token ID | 465 |
Feature activation | +7.44e-02 |
Pos logit contributions | |
on | +0.012 |
there | +0.012 |
together | +0.012 |
it | +0.011 |
in | +0.011 |
Neg logit contributions | |
ey | -0.017 |
former | -0.016 |
hotel | -0.016 |
ostr | -0.016 |
priest | -0.016 |
Token | room |
Token ID | 2119 |
Feature activation | -5.97e-02 |
Pos logit contributions | |
ey | +0.115 |
Ta | +0.105 |
priest | +0.101 |
ostr | +0.100 |
former | +0.099 |
Neg logit contributions | |
on | -0.163 |
there | -0.159 |
together | -0.158 |
es | -0.157 |
with | -0.156 |
Token | to |
Token ID | 284 |
Feature activation | +9.96e-03 |
Pos logit contributions | |
es | +0.057 |
icked | +0.049 |
on | +0.044 |
together | +0.042 |
there | +0.041 |
Neg logit contributions | |
hotel | -0.164 |
classroom | -0.160 |
city | -0.159 |
principal | -0.159 |
playground | -0.157 |
Token | get |
Token ID | 651 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
ey | +0.020 |
ostr | +0.020 |
hotel | +0.020 |
tant | +0.020 |
former | +0.020 |
Neg logit contributions | |
on | -0.016 |
there | -0.016 |
in | -0.015 |
with | -0.015 |
it | -0.014 |
Token | at |
Token ID | 379 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
es | +0.048 |
icked | +0.042 |
on | +0.031 |
together | +0.031 |
there | +0.030 |
Neg logit contributions | |
hotel | -0.118 |
city | -0.114 |
classroom | -0.113 |
museum | -0.112 |
ey | -0.112 |
Token | and |
Token ID | 290 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
es | +0.020 |
icked | +0.020 |
together | +0.015 |
on | +0.015 |
there | +0.014 |
Neg logit contributions | |
hotel | -0.050 |
city | -0.049 |
museum | -0.048 |
classroom | -0.047 |
mall | -0.047 |
Token | had |
Token ID | 550 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
es | +0.036 |
on | +0.033 |
together | +0.032 |
there | +0.032 |
icked | +0.031 |
Neg logit contributions | |
hotel | -0.060 |
ey | -0.059 |
principal | -0.058 |
city | -0.057 |
classroom | -0.057 |
Token | a |
Token ID | 257 |
Feature activation | +4.03e-02 |
Pos logit contributions | |
es | +0.023 |
on | +0.020 |
icked | +0.020 |
together | +0.020 |
there | +0.020 |
Neg logit contributions | |
hotel | -0.040 |
ey | -0.039 |
city | -0.038 |
classroom | -0.038 |
principal | -0.038 |
Token | great |
Token ID | 1049 |
Feature activation | +6.41e-02 |
Pos logit contributions | |
opl | +0.067 |
ey | +0.067 |
Ta | +0.064 |
principal | +0.064 |
orchestra | +0.063 |
Neg logit contributions | |
on | -0.081 |
there | -0.080 |
together | -0.076 |
in | -0.073 |
, | -0.071 |
Token | time |
Token ID | 640 |
Feature activation | -4.19e-02 |
Pos logit contributions | |
principal | +0.113 |
ey | +0.109 |
Ta | +0.106 |
orchestra | +0.105 |
priest | +0.104 |
Neg logit contributions | |
on | -0.123 |
there | -0.118 |
together | -0.115 |
in | -0.112 |
with | -0.110 |
Token | . |
Token ID | 13 |
Feature activation | -5.50e-03 |
Pos logit contributions | |
es | +0.042 |
icked | +0.041 |
on | +0.030 |
together | +0.030 |
there | +0.029 |
Neg logit contributions | |
hotel | -0.107 |
city | -0.106 |
classroom | -0.105 |
playground | -0.105 |
museum | -0.104 |
Token |
|
Token ID | 198 |
Feature activation | +2.33e-03 |
Pos logit contributions | |
es | +0.007 |
on | +0.006 |
there | +0.006 |
together | +0.006 |
icked | +0.006 |
Neg logit contributions | |
hotel | -0.013 |
ey | -0.013 |
principal | -0.013 |
orchestra | -0.013 |
classroom | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -2.27e-03 |
Pos logit contributions | |
hotel | +0.006 |
ey | +0.006 |
classroom | +0.006 |
city | +0.006 |
principal | +0.006 |
Neg logit contributions | |
es | -0.003 |
on | -0.002 |
together | -0.002 |
there | -0.002 |
icked | -0.002 |
Token | Suddenly |
Token ID | 38582 |
Feature activation | -3.14e-02 |
Pos logit contributions | |
es | +0.003 |
on | +0.003 |
there | +0.003 |
together | +0.003 |
icked | +0.003 |
Neg logit contributions | |
hotel | -0.005 |
ey | -0.005 |
classroom | -0.005 |
principal | -0.005 |
city | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
es | +0.029 |
icked | +0.027 |
together | +0.023 |
on | +0.020 |
instead | +0.019 |
Neg logit contributions | |
hotel | -0.088 |
city | -0.083 |
museum | -0.082 |
beach | -0.082 |
airport | -0.081 |