Token | that |
Token ID | 326 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
tted | +0.085 |
construct | +0.083 |
deserve | +0.082 |
respect | +0.082 |
rebuild | +0.081 |
Neg logit contributions | |
ittens | -0.062 |
ls | -0.059 |
ogging | -0.048 |
ule | -0.048 |
ip | -0.048 |
Token | the |
Token ID | 262 |
Feature activation | +3.96e+00 |
Pos logit contributions | |
deserve | +0.069 |
make | +0.068 |
construct | +0.068 |
respect | +0.068 |
tted | +0.067 |
Neg logit contributions | |
ittens | -0.047 |
ls | -0.041 |
ogging | -0.034 |
ip | -0.033 |
ule | -0.032 |
Token | penny |
Token ID | 26369 |
Feature activation | -2.94e+00 |
Pos logit contributions | |
tted | +0.146 |
deserve | +0.143 |
respect | +0.136 |
construct | +0.134 |
will | +0.132 |
Neg logit contributions | |
ittens | -0.074 |
ls | -0.072 |
ogging | -0.057 |
ip | -0.056 |
ule | -0.053 |
Token | find |
Token ID | 1064 |
Feature activation | +4.60e+00 |
Pos logit contributions | |
ittens | +0.062 |
ls | +0.058 |
ogging | +0.048 |
shrimp | +0.047 |
ip | +0.047 |
Neg logit contributions | |
tted | -0.100 |
deserve | -0.096 |
construct | -0.095 |
rebuild | -0.092 |
respect | -0.090 |
Token | its |
Token ID | 663 |
Feature activation | +3.19e+00 |
Pos logit contributions | |
tted | +0.149 |
construct | +0.142 |
deserve | +0.139 |
rebuild | +0.135 |
ws | +0.134 |
Neg logit contributions | |
ls | -0.103 |
ittens | -0.099 |
she | -0.094 |
strawberries | -0.085 |
shrimp | -0.084 |
Token | rightful |
Token ID | 43133 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
tted | +0.086 |
deserve | +0.081 |
construct | +0.080 |
rebuild | +0.076 |
ws | +0.075 |
Neg logit contributions | |
ittens | -0.091 |
ls | -0.088 |
ogging | -0.076 |
ip | -0.076 |
shrimp | -0.075 |
Token | owner |
Token ID | 4870 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
tted | +0.335 |
construct | +0.328 |
rebuild | +0.320 |
deserve | +0.309 |
ws | +0.298 |
Neg logit contributions | |
ittens | -0.208 |
ls | -0.203 |
strawberries | -0.189 |
strawberry | -0.180 |
she | -0.178 |
Token | . |
Token ID | 13 |
Feature activation | +5.07e-01 |
Pos logit contributions | |
tted | +0.093 |
deserve | +0.089 |
construct | +0.089 |
ws | +0.086 |
rebuild | +0.086 |
Neg logit contributions | |
ittens | -0.035 |
ls | -0.033 |
she | -0.025 |
ip | -0.025 |
shrimp | -0.024 |
Token | |
Token ID | 220 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
tted | +0.013 |
deserve | +0.012 |
construct | +0.011 |
rebuild | +0.011 |
overcome | +0.010 |
Neg logit contributions | |
ls | -0.016 |
ittens | -0.016 |
strawberries | -0.013 |
shrimp | -0.013 |
ule | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | +3.03e+00 |
Pos logit contributions | |
ittens | +0.040 |
ls | +0.037 |
ip | +0.032 |
she | +0.032 |
ogging | +0.031 |
Neg logit contributions | |
tted | -0.052 |
deserve | -0.050 |
construct | -0.050 |
rebuild | -0.048 |
ws | -0.048 |
Token |
|
Token ID | 198 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
tted | +0.091 |
deserve | +0.086 |
construct | +0.086 |
ws | +0.083 |
rebuild | +0.083 |
Neg logit contributions | |
ittens | -0.077 |
ls | -0.072 |
ogging | -0.062 |
she | -0.062 |
ip | -0.062 |
Token | to |
Token ID | 284 |
Feature activation | +3.71e+00 |
Pos logit contributions | |
tted | +0.149 |
deserve | +0.142 |
construct | +0.140 |
rebuild | +0.136 |
ws | +0.134 |
Neg logit contributions | |
ittens | -0.095 |
ls | -0.090 |
shrimp | -0.074 |
ogging | -0.074 |
ip | -0.073 |
Token | return |
Token ID | 1441 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
tted | +0.078 |
construct | +0.066 |
deserve | +0.063 |
ws | +0.062 |
rebuild | +0.059 |
Neg logit contributions | |
ittens | -0.126 |
ls | -0.117 |
she | -0.115 |
shrimp | -0.114 |
Gig | -0.112 |
Token | it |
Token ID | 340 |
Feature activation | +2.59e+00 |
Pos logit contributions | |
tted | +0.041 |
construct | +0.039 |
deserve | +0.038 |
rebuild | +0.037 |
ws | +0.037 |
Neg logit contributions | |
ittens | -0.020 |
ls | -0.019 |
shrimp | -0.015 |
strawberries | -0.015 |
ip | -0.015 |
Token | to |
Token ID | 284 |
Feature activation | +4.43e+00 |
Pos logit contributions | |
tted | +0.081 |
construct | +0.077 |
deserve | +0.076 |
rebuild | +0.074 |
ws | +0.073 |
Neg logit contributions | |
ittens | -0.060 |
ls | -0.059 |
she | -0.049 |
shrimp | -0.049 |
ip | -0.049 |
Token | its |
Token ID | 663 |
Feature activation | +3.29e+00 |
Pos logit contributions | |
tted | +0.150 |
construct | +0.133 |
deserve | +0.129 |
ws | +0.127 |
rebuild | +0.124 |
Neg logit contributions | |
ittens | -0.098 |
ls | -0.093 |
she | -0.087 |
shrimp | -0.085 |
strawberries | -0.082 |
Token | rightful |
Token ID | 43133 |
Feature activation | +1.00e+01 |
Pos logit contributions | |
tted | +0.083 |
deserve | +0.076 |
construct | +0.075 |
rebuild | +0.071 |
ws | +0.071 |
Neg logit contributions | |
ittens | -0.100 |
ls | -0.097 |
shrimp | -0.085 |
strawberries | -0.085 |
ip | -0.084 |
Token | owner |
Token ID | 4870 |
Feature activation | +2.57e+00 |
Pos logit contributions | |
tted | +0.334 |
construct | +0.321 |
rebuild | +0.312 |
deserve | +0.302 |
ws | +0.287 |
Neg logit contributions | |
ls | -0.193 |
ittens | -0.191 |
strawberries | -0.179 |
- | -0.171 |
strawberry | -0.171 |
Token | if |
Token ID | 611 |
Feature activation | -6.49e-01 |
Pos logit contributions | |
tted | +0.085 |
construct | +0.080 |
deserve | +0.079 |
rebuild | +0.077 |
ws | +0.077 |
Neg logit contributions | |
ittens | -0.056 |
ls | -0.054 |
ip | -0.046 |
she | -0.045 |
ogging | -0.045 |
Token | you |
Token ID | 345 |
Feature activation | +3.63e+00 |
Pos logit contributions | |
ittens | +0.013 |
ls | +0.013 |
ogging | +0.010 |
ip | +0.010 |
shrimp | +0.010 |
Neg logit contributions | |
tted | -0.023 |
deserve | -0.022 |
construct | -0.021 |
rebuild | -0.021 |
ws | -0.020 |
Token | can |
Token ID | 460 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
tted | +0.074 |
construct | +0.070 |
stations | +0.063 |
rebuild | +0.063 |
ws | +0.061 |
Neg logit contributions | |
ls | -0.116 |
ittens | -0.115 |
shrimp | -0.108 |
she | -0.107 |
ogging | -0.105 |
Token | find |
Token ID | 1064 |
Feature activation | +5.25e+00 |
Pos logit contributions | |
tted | +0.045 |
ws | +0.043 |
construct | +0.043 |
stations | +0.041 |
deserve | +0.041 |
Neg logit contributions | |
ittens | -0.069 |
ls | -0.065 |
she | -0.061 |
ogging | -0.060 |
ip | -0.059 |
Token | returned |
Token ID | 4504 |
Feature activation | +7.59e-01 |
Pos logit contributions | |
ls | +0.211 |
ittens | +0.211 |
she | +0.208 |
shrimp | +0.193 |
he | +0.188 |
Neg logit contributions | |
tted | -0.213 |
construct | -0.192 |
ws | -0.191 |
deserve | -0.189 |
rebuild | -0.176 |
Token | the |
Token ID | 262 |
Feature activation | +4.19e+00 |
Pos logit contributions | |
tted | +0.022 |
deserve | +0.021 |
construct | +0.021 |
rebuild | +0.020 |
respect | +0.020 |
Neg logit contributions | |
ittens | -0.020 |
ls | -0.019 |
ule | -0.017 |
ogging | -0.016 |
ip | -0.016 |
Token | towel |
Token ID | 24808 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
tted | +0.129 |
deserve | +0.122 |
construct | +0.119 |
respect | +0.115 |
rebuild | +0.114 |
Neg logit contributions | |
ittens | -0.104 |
ls | -0.098 |
ip | -0.085 |
ule | -0.084 |
ogging | -0.083 |
Token | to |
Token ID | 284 |
Feature activation | +3.85e+00 |
Pos logit contributions | |
ittens | +0.046 |
ls | +0.046 |
she | +0.039 |
ip | +0.038 |
shrimp | +0.038 |
Neg logit contributions | |
tted | -0.069 |
deserve | -0.067 |
construct | -0.066 |
ws | -0.063 |
rebuild | -0.063 |
Token | its |
Token ID | 663 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
tted | +0.144 |
deserve | +0.135 |
construct | +0.133 |
ws | +0.129 |
rebuild | +0.126 |
Neg logit contributions | |
ittens | -0.070 |
ls | -0.065 |
she | -0.058 |
shrimp | -0.055 |
ogging | -0.052 |
Token | rightful |
Token ID | 43133 |
Feature activation | +9.88e+00 |
Pos logit contributions | |
tted | +0.073 |
construct | +0.070 |
deserve | +0.069 |
respect | +0.068 |
rebuild | +0.068 |
Neg logit contributions | |
ittens | -0.078 |
ls | -0.075 |
ogging | -0.067 |
ip | -0.066 |
ipop | -0.065 |
Token | owner |
Token ID | 4870 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
tted | +0.334 |
construct | +0.324 |
deserve | +0.313 |
rebuild | +0.312 |
ws | +0.297 |
Neg logit contributions | |
ls | -0.194 |
ittens | -0.193 |
strawberries | -0.173 |
shrimp | -0.167 |
strawberry | -0.164 |
Token | . |
Token ID | 13 |
Feature activation | -3.96e-01 |
Pos logit contributions | |
tted | +0.078 |
deserve | +0.074 |
construct | +0.073 |
rebuild | +0.071 |
ws | +0.070 |
Neg logit contributions | |
ittens | -0.045 |
ls | -0.043 |
shrimp | -0.035 |
ogging | -0.035 |
ip | -0.034 |
Token | The |
Token ID | 383 |
Feature activation | +2.87e+00 |
Pos logit contributions | |
ittens | +0.010 |
ls | +0.010 |
shrimp | +0.009 |
strawberries | +0.009 |
ule | +0.009 |
Neg logit contributions | |
tted | -0.012 |
deserve | -0.011 |
construct | -0.011 |
rebuild | -0.010 |
respect | -0.010 |
Token | owner |
Token ID | 4870 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
tted | +0.065 |
deserve | +0.060 |
construct | +0.057 |
respect | +0.056 |
rebuild | +0.055 |
Neg logit contributions | |
ittens | -0.096 |
ls | -0.093 |
ogging | -0.083 |
ip | -0.083 |
ule | -0.082 |
Token | was |
Token ID | 373 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
tted | +0.044 |
deserve | +0.042 |
construct | +0.041 |
ws | +0.040 |
rebuild | +0.039 |
Neg logit contributions | |
ittens | -0.033 |
ls | -0.032 |
she | -0.027 |
ip | -0.027 |
shrimp | -0.027 |
Token | sign |
Token ID | 1051 |
Feature activation | -8.85e-01 |
Pos logit contributions | |
tted | +0.005 |
deserve | +0.005 |
construct | +0.005 |
make | +0.005 |
will | +0.005 |
Neg logit contributions | |
ittens | -0.005 |
ls | -0.005 |
ogging | -0.004 |
ries | -0.004 |
ip | -0.004 |
Token | that |
Token ID | 326 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
ittens | +0.018 |
ls | +0.015 |
ip | +0.013 |
ule | +0.013 |
ogging | +0.013 |
Neg logit contributions | |
tted | -0.032 |
deserve | -0.030 |
construct | -0.030 |
respect | -0.029 |
rebuild | -0.029 |
Token | says |
Token ID | 1139 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
tted | +0.045 |
deserve | +0.042 |
construct | +0.042 |
rebuild | +0.041 |
ws | +0.040 |
Neg logit contributions | |
ls | -0.032 |
ittens | -0.032 |
she | -0.027 |
shrimp | -0.026 |
ip | -0.026 |
Token | " |
Token ID | 366 |
Feature activation | +3.64e+00 |
Pos logit contributions | |
ittens | +0.003 |
ls | +0.003 |
shrimp | +0.002 |
ip | +0.002 |
ogging | +0.002 |
Neg logit contributions | |
tted | -0.006 |
deserve | -0.005 |
construct | -0.005 |
rebuild | -0.005 |
ws | -0.005 |
Token | Ad |
Token ID | 2782 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
construct | +0.089 |
tted | +0.088 |
rebuild | +0.086 |
ws | +0.084 |
overcome | +0.081 |
Neg logit contributions | |
ittens | -0.105 |
ls | -0.101 |
she | -0.091 |
ogging | -0.090 |
ip | -0.089 |
Token | opt |
Token ID | 8738 |
Feature activation | +9.74e+00 |
Pos logit contributions | |
ittens | +0.063 |
ls | +0.061 |
ip | +0.052 |
ogging | +0.050 |
ule | +0.050 |
Neg logit contributions | |
tted | -0.087 |
construct | -0.084 |
deserve | -0.082 |
rebuild | -0.081 |
ws | -0.079 |
Token | a |
Token ID | 257 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
tted | +0.293 |
construct | +0.275 |
deserve | +0.271 |
rebuild | +0.264 |
ws | +0.260 |
Neg logit contributions | |
ittens | -0.238 |
ls | -0.237 |
shrimp | -0.209 |
strawberries | -0.203 |
Gig | -0.200 |
Token | Panda |
Token ID | 41112 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
tted | +0.043 |
deserve | +0.042 |
construct | +0.041 |
respect | +0.039 |
make | +0.038 |
Neg logit contributions | |
ittens | -0.037 |
ls | -0.035 |
ogging | -0.030 |
ries | -0.030 |
ip | -0.029 |
Token | ". |
Token ID | 1911 |
Feature activation | +6.69e-01 |
Pos logit contributions | |
ittens | +0.063 |
ls | +0.061 |
ogging | +0.052 |
strawberries | +0.051 |
shrimp | +0.051 |
Neg logit contributions | |
tted | -0.069 |
deserve | -0.065 |
construct | -0.063 |
rebuild | -0.060 |
respect | -0.060 |
Token | They |
Token ID | 1119 |
Feature activation | +3.09e+00 |
Pos logit contributions | |
tted | +0.018 |
deserve | +0.017 |
construct | +0.016 |
rebuild | +0.016 |
ws | +0.015 |
Neg logit contributions | |
ittens | -0.020 |
ls | -0.019 |
ogging | -0.016 |
ule | -0.016 |
ip | -0.016 |
Token | ask |
Token ID | 1265 |
Feature activation | +3.04e+00 |
Pos logit contributions | |
tted | +0.074 |
construct | +0.060 |
deserve | +0.059 |
ction | +0.059 |
ws | +0.059 |
Neg logit contributions | |
sc | -0.090 |
ls | -0.087 |
she | -0.086 |
ittens | -0.085 |
wo | -0.082 |
Token | the |
Token ID | 262 |
Feature activation | +3.66e+00 |
Pos logit contributions | |
tted | +0.005 |
deserve | +0.004 |
construct | +0.004 |
rebuild | +0.004 |
respect | +0.004 |
Neg logit contributions | |
ittens | -0.003 |
ls | -0.003 |
ule | -0.003 |
ogging | -0.003 |
ip | -0.003 |
Token | stolen |
Token ID | 9909 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
tted | +0.108 |
deserve | +0.102 |
construct | +0.102 |
ws | +0.098 |
rebuild | +0.098 |
Neg logit contributions | |
ittens | -0.093 |
ls | -0.091 |
shrimp | -0.080 |
strawberries | -0.078 |
she | -0.077 |
Token | squares |
Token ID | 24438 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
tted | +0.034 |
deserve | +0.033 |
construct | +0.033 |
rebuild | +0.032 |
ws | +0.031 |
Neg logit contributions | |
ls | -0.027 |
ittens | -0.025 |
shrimp | -0.024 |
strawberries | -0.023 |
she | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | +3.96e+00 |
Pos logit contributions | |
tted | +0.068 |
deserve | +0.063 |
construct | +0.062 |
rebuild | +0.061 |
ws | +0.060 |
Neg logit contributions | |
ittens | -0.046 |
ls | -0.042 |
ule | -0.035 |
ip | -0.035 |
ogging | -0.034 |
Token | their |
Token ID | 511 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
tted | +0.142 |
deserve | +0.132 |
construct | +0.130 |
ws | +0.129 |
gen | +0.122 |
Neg logit contributions | |
ls | -0.070 |
ittens | -0.070 |
shrimp | -0.064 |
she | -0.064 |
Gig | -0.061 |
Token | rightful |
Token ID | 43133 |
Feature activation | +9.65e+00 |
Pos logit contributions | |
tted | +0.056 |
construct | +0.053 |
rebuild | +0.053 |
deserve | +0.052 |
respect | +0.051 |
Neg logit contributions | |
ittens | -0.055 |
ls | -0.050 |
ogging | -0.045 |
ip | -0.044 |
ule | -0.044 |
Token | place |
Token ID | 1295 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
tted | +0.328 |
construct | +0.318 |
rebuild | +0.312 |
deserve | +0.306 |
ws | +0.291 |
Neg logit contributions | |
ls | -0.182 |
ittens | -0.176 |
strawberries | -0.166 |
shrimp | -0.163 |
strawberry | -0.158 |
Token | . |
Token ID | 13 |
Feature activation | +9.65e-02 |
Pos logit contributions | |
ittens | +0.024 |
ls | +0.023 |
shrimp | +0.019 |
ip | +0.019 |
ogging | +0.019 |
Neg logit contributions | |
tted | -0.040 |
deserve | -0.038 |
construct | -0.037 |
rebuild | -0.036 |
ws | -0.036 |
Token |
|
Token ID | 198 |
Feature activation | +3.03e+00 |
Pos logit contributions | |
tted | +0.003 |
deserve | +0.002 |
construct | +0.002 |
rebuild | +0.002 |
respect | +0.002 |
Neg logit contributions | |
ittens | -0.003 |
ls | -0.003 |
strawberries | -0.002 |
ogging | -0.002 |
shrimp | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
tted | +0.088 |
deserve | +0.083 |
construct | +0.082 |
rebuild | +0.079 |
ws | +0.078 |
Neg logit contributions | |
ittens | -0.080 |
ls | -0.077 |
ogging | -0.065 |
ip | -0.065 |
shrimp | -0.065 |
Token | The |
Token ID | 464 |
Feature activation | +2.65e+00 |
Pos logit contributions | |
tted | +0.075 |
construct | +0.074 |
deserve | +0.073 |
ws | +0.072 |
rebuild | +0.071 |
Neg logit contributions | |
ittens | -0.060 |
ls | -0.056 |
ule | -0.048 |
shrimp | -0.047 |
she | -0.047 |
Token | put |
Token ID | 1234 |
Feature activation | +8.59e-01 |
Pos logit contributions | |
ittens | +0.244 |
ls | +0.243 |
she | +0.233 |
shrimp | +0.222 |
strawberries | +0.219 |
Neg logit contributions | |
tted | -0.157 |
deserve | -0.137 |
construct | -0.130 |
ws | -0.129 |
gen | -0.124 |
Token | them |
Token ID | 606 |
Feature activation | +3.71e+00 |
Pos logit contributions | |
tted | +0.027 |
deserve | +0.025 |
construct | +0.025 |
rebuild | +0.024 |
ws | +0.024 |
Neg logit contributions | |
ittens | -0.021 |
ls | -0.020 |
shrimp | -0.017 |
ip | -0.017 |
strawberries | -0.017 |
Token | back |
Token ID | 736 |
Feature activation | -3.03e-02 |
Pos logit contributions | |
tted | +0.110 |
deserve | +0.101 |
construct | +0.100 |
rebuild | +0.097 |
ws | +0.096 |
Neg logit contributions | |
ittens | -0.098 |
ls | -0.092 |
ule | -0.083 |
ogging | -0.081 |
shrimp | -0.080 |
Token | in |
Token ID | 287 |
Feature activation | +4.70e+00 |
Pos logit contributions | |
ittens | +0.001 |
ls | +0.001 |
ule | +0.001 |
shrimp | +0.001 |
ogging | +0.001 |
Neg logit contributions | |
tted | -0.001 |
deserve | -0.001 |
construct | -0.001 |
rebuild | -0.001 |
ws | -0.001 |
Token | their |
Token ID | 511 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
tted | +0.159 |
deserve | +0.150 |
construct | +0.146 |
ws | +0.142 |
rebuild | +0.140 |
Neg logit contributions | |
ittens | -0.098 |
ls | -0.093 |
shrimp | -0.083 |
she | -0.081 |
strawberries | -0.080 |
Token | rightful |
Token ID | 43133 |
Feature activation | +9.64e+00 |
Pos logit contributions | |
tted | +0.060 |
deserve | +0.057 |
construct | +0.056 |
rebuild | +0.054 |
ws | +0.053 |
Neg logit contributions | |
ittens | -0.057 |
ls | -0.055 |
ule | -0.048 |
ogging | -0.048 |
ip | -0.047 |
Token | place |
Token ID | 1295 |
Feature activation | -6.20e-01 |
Pos logit contributions | |
tted | +0.273 |
rebuild | +0.254 |
construct | +0.253 |
deserve | +0.249 |
affect | +0.233 |
Neg logit contributions | |
ittens | -0.235 |
ls | -0.230 |
strawberries | -0.223 |
because | -0.212 |
strawberry | -0.212 |
Token | . |
Token ID | 13 |
Feature activation | +8.91e-01 |
Pos logit contributions | |
ittens | +0.013 |
ls | +0.012 |
shrimp | +0.010 |
ogging | +0.010 |
ip | +0.010 |
Neg logit contributions | |
tted | -0.021 |
deserve | -0.020 |
construct | -0.020 |
rebuild | -0.019 |
ws | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | +3.42e+00 |
Pos logit contributions | |
tted | +0.027 |
deserve | +0.025 |
construct | +0.025 |
ws | +0.024 |
rebuild | +0.024 |
Neg logit contributions | |
ittens | -0.023 |
ls | -0.021 |
ogging | -0.019 |
she | -0.018 |
shrimp | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | +2.78e+00 |
Pos logit contributions | |
tted | +0.103 |
deserve | +0.098 |
construct | +0.097 |
rebuild | +0.093 |
ws | +0.093 |
Neg logit contributions | |
ittens | -0.087 |
ls | -0.083 |
ogging | -0.071 |
ip | -0.070 |
ule | -0.070 |
Token | After |
Token ID | 3260 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
tted | +0.081 |
ws | +0.079 |
construct | +0.079 |
deserve | +0.078 |
rebuild | +0.076 |
Neg logit contributions | |
ittens | -0.070 |
ls | -0.063 |
ogging | -0.058 |
she | -0.058 |
shrimp | -0.057 |
Token | the |
Token ID | 262 |
Feature activation | +3.44e+00 |
Pos logit contributions | |
ittens | +0.204 |
ls | +0.190 |
ogging | +0.155 |
ule | +0.154 |
ip | +0.148 |
Neg logit contributions | |
deserve | -0.323 |
tted | -0.323 |
construct | -0.309 |
rebuild | -0.306 |
respect | -0.306 |
Token | ancient |
Token ID | 6156 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
tted | +0.107 |
deserve | +0.103 |
construct | +0.100 |
respect | +0.098 |
rebuild | +0.097 |
Neg logit contributions | |
ittens | -0.081 |
ls | -0.077 |
ogging | -0.066 |
ip | -0.064 |
ries | -0.063 |
Token | knob |
Token ID | 36842 |
Feature activation | -2.67e+00 |
Pos logit contributions | |
ittens | +0.027 |
ls | +0.024 |
ogging | +0.021 |
ries | +0.021 |
ip | +0.021 |
Neg logit contributions | |
tted | -0.029 |
deserve | -0.028 |
respect | -0.028 |
construct | -0.027 |
make | -0.026 |
Token | found |
Token ID | 1043 |
Feature activation | +2.75e+00 |
Pos logit contributions | |
ls | +0.066 |
ittens | +0.064 |
she | +0.060 |
ip | +0.054 |
ipop | +0.054 |
Neg logit contributions | |
tted | -0.079 |
deserve | -0.076 |
construct | -0.075 |
ws | -0.072 |
rebuild | -0.070 |
Token | its |
Token ID | 663 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
tted | +0.088 |
deserve | +0.085 |
construct | +0.084 |
ws | +0.081 |
rebuild | +0.080 |
Neg logit contributions | |
ls | -0.061 |
ittens | -0.059 |
she | -0.054 |
shrimp | -0.050 |
strawberries | -0.050 |
Token | rightful |
Token ID | 43133 |
Feature activation | +9.63e+00 |
Pos logit contributions | |
tted | +0.074 |
deserve | +0.069 |
construct | +0.068 |
ws | +0.066 |
rebuild | +0.065 |
Neg logit contributions | |
ittens | -0.063 |
ls | -0.060 |
shrimp | -0.052 |
strawberries | -0.052 |
she | -0.051 |
Token | place |
Token ID | 1295 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
tted | +0.329 |
construct | +0.321 |
rebuild | +0.317 |
deserve | +0.307 |
ws | +0.295 |
Neg logit contributions | |
ittens | -0.187 |
ls | -0.170 |
strawberries | -0.161 |
she | -0.158 |
strawberry | -0.157 |
Token | in |
Token ID | 287 |
Feature activation | +3.75e+00 |
Pos logit contributions | |
ittens | +0.028 |
ls | +0.026 |
shrimp | +0.020 |
ogging | +0.020 |
ip | +0.020 |
Neg logit contributions | |
tted | -0.062 |
deserve | -0.059 |
construct | -0.059 |
rebuild | -0.057 |
ws | -0.056 |
Token | their |
Token ID | 511 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
tted | +0.125 |
deserve | +0.114 |
construct | +0.113 |
ws | +0.111 |
rebuild | +0.109 |
Neg logit contributions | |
ittens | -0.077 |
ls | -0.076 |
she | -0.072 |
shrimp | -0.069 |
hungry | -0.066 |
Token | house |
Token ID | 2156 |
Feature activation | -8.25e-01 |
Pos logit contributions | |
tted | +0.044 |
deserve | +0.041 |
construct | +0.040 |
rebuild | +0.038 |
ws | +0.038 |
Neg logit contributions | |
ittens | -0.044 |
ls | -0.043 |
shrimp | -0.037 |
strawberries | -0.037 |
ogging | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
ittens | +0.014 |
ls | +0.013 |
she | +0.011 |
ogging | +0.011 |
ip | +0.011 |
Neg logit contributions | |
tted | -0.031 |
deserve | -0.030 |
construct | -0.029 |
ws | -0.029 |
rebuild | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | +3.13e+00 |
Pos logit contributions | |
tted | +0.030 |
deserve | +0.028 |
construct | +0.028 |
rebuild | +0.027 |
ws | +0.026 |
Neg logit contributions | |
ittens | -0.023 |
ls | -0.022 |
ogging | -0.019 |
ule | -0.018 |
shrimp | -0.018 |
Token | rice |
Token ID | 11464 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
tted | +0.102 |
deserve | +0.097 |
construct | +0.096 |
rebuild | +0.092 |
ws | +0.092 |
Neg logit contributions | |
ittens | -0.071 |
ls | -0.068 |
shrimp | -0.059 |
strawberries | -0.057 |
ule | -0.056 |
Token | back |
Token ID | 736 |
Feature activation | -8.09e-01 |
Pos logit contributions | |
ittens | +0.027 |
ls | +0.025 |
ule | +0.022 |
ogging | +0.022 |
ip | +0.021 |
Neg logit contributions | |
tted | -0.032 |
deserve | -0.030 |
construct | -0.030 |
rebuild | -0.029 |
ws | -0.029 |
Token | in |
Token ID | 287 |
Feature activation | +4.13e+00 |
Pos logit contributions | |
ittens | +0.020 |
ls | +0.018 |
ule | +0.016 |
ogging | +0.015 |
shrimp | +0.015 |
Neg logit contributions | |
tted | -0.025 |
deserve | -0.023 |
construct | -0.023 |
rebuild | -0.023 |
ws | -0.023 |
Token | its |
Token ID | 663 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
tted | +0.140 |
deserve | +0.131 |
construct | +0.128 |
ws | +0.124 |
rebuild | +0.122 |
Neg logit contributions | |
ittens | -0.087 |
ls | -0.083 |
shrimp | -0.076 |
strawberries | -0.073 |
she | -0.073 |
Token | rightful |
Token ID | 43133 |
Feature activation | +9.60e+00 |
Pos logit contributions | |
tted | +0.068 |
deserve | +0.065 |
construct | +0.064 |
rebuild | +0.061 |
ws | +0.061 |
Neg logit contributions | |
ittens | -0.055 |
ls | -0.053 |
shrimp | -0.045 |
strawberries | -0.044 |
ip | -0.044 |
Token | place |
Token ID | 1295 |
Feature activation | -8.72e-01 |
Pos logit contributions | |
tted | +0.287 |
construct | +0.273 |
rebuild | +0.266 |
deserve | +0.262 |
affect | +0.245 |
Neg logit contributions | |
ls | -0.217 |
ittens | -0.216 |
strawberries | -0.211 |
strawberry | -0.207 |
shrimp | -0.204 |
Token | . |
Token ID | 13 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
ittens | +0.018 |
ls | +0.017 |
ogging | +0.014 |
shrimp | +0.014 |
ip | +0.014 |
Neg logit contributions | |
tted | -0.030 |
deserve | -0.029 |
construct | -0.028 |
rebuild | -0.028 |
respect | -0.027 |
Token | John |
Token ID | 1757 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
tted | +0.006 |
construct | +0.006 |
deserve | +0.006 |
ws | +0.005 |
rebuild | +0.005 |
Neg logit contributions | |
ittens | -0.004 |
ls | -0.004 |
she | -0.004 |
ogging | -0.004 |
ule | -0.004 |
Token | was |
Token ID | 373 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
ittens | +0.009 |
ls | +0.008 |
ip | +0.007 |
shrimp | +0.007 |
ule | +0.007 |
Neg logit contributions | |
tted | -0.011 |
deserve | -0.010 |
construct | -0.010 |
rebuild | -0.009 |
ws | -0.009 |
Token | proud |
Token ID | 6613 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
ittens | +0.027 |
ls | +0.025 |
she | +0.021 |
hungry | +0.020 |
shrimp | +0.020 |
Neg logit contributions | |
tted | -0.068 |
construct | -0.065 |
deserve | -0.065 |
ws | -0.064 |
rebuild | -0.063 |
Token | restore |
Token ID | 11169 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
ittens | +0.198 |
ls | +0.197 |
she | +0.183 |
shrimp | +0.180 |
strawberries | +0.175 |
Neg logit contributions | |
tted | -0.227 |
deserve | -0.203 |
construct | -0.196 |
ws | -0.194 |
rebuild | -0.185 |
Token | it |
Token ID | 340 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
tted | +0.047 |
deserve | +0.044 |
construct | +0.043 |
rebuild | +0.041 |
ws | +0.041 |
Neg logit contributions | |
ls | -0.029 |
ittens | -0.028 |
shrimp | -0.024 |
strawberries | -0.023 |
she | -0.023 |
Token | back |
Token ID | 736 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
tted | +0.064 |
deserve | +0.061 |
construct | +0.060 |
ws | +0.058 |
rebuild | +0.057 |
Neg logit contributions | |
ls | -0.031 |
ittens | -0.029 |
she | -0.026 |
shrimp | -0.023 |
ip | -0.023 |
Token | to |
Token ID | 284 |
Feature activation | +3.53e+00 |
Pos logit contributions | |
ittens | +0.009 |
ls | +0.009 |
she | +0.008 |
shrimp | +0.008 |
ogging | +0.007 |
Neg logit contributions | |
tted | -0.017 |
deserve | -0.017 |
construct | -0.016 |
ws | -0.016 |
rebuild | -0.016 |
Token | its |
Token ID | 663 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
tted | +0.142 |
deserve | +0.126 |
construct | +0.121 |
ws | +0.120 |
reed | +0.118 |
Neg logit contributions | |
ittens | -0.052 |
she | -0.049 |
ls | -0.049 |
shrimp | -0.048 |
he | -0.047 |
Token | rightful |
Token ID | 43133 |
Feature activation | +9.56e+00 |
Pos logit contributions | |
tted | +0.074 |
deserve | +0.069 |
construct | +0.068 |
ws | +0.065 |
rebuild | +0.064 |
Neg logit contributions | |
ittens | -0.073 |
ls | -0.071 |
shrimp | -0.062 |
strawberries | -0.061 |
she | -0.061 |
Token | place |
Token ID | 1295 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
tted | +0.299 |
construct | +0.282 |
rebuild | +0.277 |
deserve | +0.272 |
ws | +0.254 |
Neg logit contributions | |
ls | -0.203 |
ittens | -0.199 |
strawberries | -0.196 |
strawberry | -0.186 |
because | -0.185 |
Token | . |
Token ID | 13 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
ittens | +0.018 |
ls | +0.018 |
she | +0.015 |
shrimp | +0.014 |
ogging | +0.014 |
Neg logit contributions | |
tted | -0.039 |
deserve | -0.037 |
construct | -0.037 |
ws | -0.036 |
rebuild | -0.035 |
Token | Although |
Token ID | 4900 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
tted | +0.008 |
construct | +0.008 |
deserve | +0.007 |
ws | +0.007 |
rebuild | +0.007 |
Neg logit contributions | |
ittens | -0.005 |
ls | -0.005 |
ip | -0.005 |
she | -0.004 |
ogging | -0.004 |
Token | Sammy |
Token ID | 44067 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
ittens | +0.023 |
ls | +0.020 |
ipop | +0.017 |
ule | +0.017 |
ogging | +0.016 |
Neg logit contributions | |
tted | -0.041 |
deserve | -0.040 |
construct | -0.040 |
overcome | -0.040 |
rebuild | -0.039 |
Token | was |
Token ID | 373 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
tted | +0.005 |
deserve | +0.005 |
construct | +0.005 |
rebuild | +0.004 |
ws | +0.004 |
Neg logit contributions | |
ls | -0.005 |
ittens | -0.005 |
ip | -0.004 |
she | -0.004 |
shrimp | -0.004 |
Token | The |
Token ID | 383 |
Feature activation | +2.72e+00 |
Pos logit contributions | |
ittens | +0.015 |
ls | +0.015 |
shrimp | +0.013 |
ogging | +0.013 |
ip | +0.013 |
Neg logit contributions | |
tted | -0.012 |
deserve | -0.011 |
construct | -0.011 |
rebuild | -0.010 |
ws | -0.010 |
Token | boot |
Token ID | 6297 |
Feature activation | -3.88e+00 |
Pos logit contributions | |
tted | +0.086 |
deserve | +0.081 |
construct | +0.081 |
rebuild | +0.077 |
respect | +0.076 |
Neg logit contributions | |
ittens | -0.065 |
ls | -0.063 |
ogging | -0.052 |
ries | -0.051 |
ip | -0.051 |
Token | returned |
Token ID | 4504 |
Feature activation | +4.23e-02 |
Pos logit contributions | |
ittens | +0.110 |
ls | +0.104 |
ogging | +0.093 |
shrimp | +0.091 |
ule | +0.090 |
Neg logit contributions | |
tted | -0.109 |
deserve | -0.099 |
construct | -0.097 |
rebuild | -0.093 |
overcome | -0.092 |
Token | to |
Token ID | 284 |
Feature activation | +3.71e+00 |
Pos logit contributions | |
tted | +0.002 |
deserve | +0.001 |
construct | +0.001 |
rebuild | +0.001 |
ws | +0.001 |
Neg logit contributions | |
ittens | -0.001 |
ls | -0.001 |
ogging | -0.001 |
shrimp | -0.001 |
ip | -0.001 |
Token | its |
Token ID | 663 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
tted | +0.127 |
deserve | +0.117 |
construct | +0.115 |
ws | +0.112 |
rebuild | +0.109 |
Neg logit contributions | |
ittens | -0.076 |
ls | -0.074 |
she | -0.067 |
shrimp | -0.066 |
Gig | -0.064 |
Token | rightful |
Token ID | 43133 |
Feature activation | +9.51e+00 |
Pos logit contributions | |
tted | +0.065 |
deserve | +0.061 |
construct | +0.061 |
rebuild | +0.058 |
respect | +0.057 |
Neg logit contributions | |
ittens | -0.072 |
ls | -0.070 |
ogging | -0.061 |
ip | -0.060 |
ule | -0.060 |
Token | owners |
Token ID | 4393 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
tted | +0.315 |
construct | +0.300 |
deserve | +0.297 |
rebuild | +0.296 |
ws | +0.279 |
Neg logit contributions | |
ittens | -0.195 |
ls | -0.189 |
strawberries | -0.172 |
Gig | -0.165 |
strawberry | -0.164 |
Token | and |
Token ID | 290 |
Feature activation | -7.78e+00 |
Pos logit contributions | |
tted | +0.064 |
deserve | +0.061 |
construct | +0.060 |
rebuild | +0.058 |
ws | +0.058 |
Neg logit contributions | |
ittens | -0.022 |
ls | -0.021 |
shrimp | -0.015 |
ogging | -0.015 |
ip | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | +3.89e+00 |
Pos logit contributions | |
ittens | +0.151 |
ls | +0.148 |
she | +0.126 |
shrimp | +0.124 |
strawberries | +0.119 |
Neg logit contributions | |
tted | -0.275 |
deserve | -0.258 |
construct | -0.256 |
ws | -0.250 |
rebuild | -0.244 |
Token | conflict |
Token ID | 5358 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
tted | +0.126 |
deserve | +0.119 |
construct | +0.118 |
rebuild | +0.113 |
respect | +0.111 |
Neg logit contributions | |
ittens | -0.091 |
ls | -0.087 |
ogging | -0.072 |
ip | -0.072 |
ule | -0.071 |
Token | was |
Token ID | 373 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
tted | +0.034 |
deserve | +0.032 |
construct | +0.032 |
ws | +0.031 |
rebuild | +0.030 |
Neg logit contributions | |
ittens | -0.023 |
ls | -0.023 |
she | -0.019 |
shrimp | -0.019 |
ip | -0.019 |
Token | could |
Token ID | 714 |
Feature activation | +3.01e+00 |
Pos logit contributions | |
tted | +0.042 |
deserve | +0.040 |
construct | +0.039 |
ws | +0.037 |
rebuild | +0.037 |
Neg logit contributions | |
ls | -0.038 |
ittens | -0.038 |
she | -0.032 |
shrimp | -0.032 |
ip | -0.032 |
Token | be |
Token ID | 307 |
Feature activation | +4.30e-01 |
Pos logit contributions | |
tted | +0.060 |
ws | +0.056 |
construct | +0.055 |
deserve | +0.054 |
irteen | +0.051 |
Neg logit contributions | |
ittens | -0.100 |
ls | -0.097 |
she | -0.097 |
sc | -0.090 |
Twe | -0.089 |
Token | back |
Token ID | 736 |
Feature activation | -7.16e-01 |
Pos logit contributions | |
tted | +0.014 |
deserve | +0.013 |
construct | +0.013 |
rebuild | +0.012 |
ws | +0.012 |
Neg logit contributions | |
ittens | -0.010 |
ls | -0.010 |
ule | -0.008 |
ogging | -0.008 |
ip | -0.008 |
Token | with |
Token ID | 351 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
ittens | +0.015 |
ls | +0.015 |
shrimp | +0.012 |
ogging | +0.012 |
ip | +0.012 |
Neg logit contributions | |
tted | -0.024 |
deserve | -0.023 |
construct | -0.023 |
rebuild | -0.022 |
ws | -0.022 |
Token | its |
Token ID | 663 |
Feature activation | +2.75e+00 |
Pos logit contributions | |
tted | +0.074 |
deserve | +0.070 |
construct | +0.069 |
rebuild | +0.066 |
ws | +0.066 |
Neg logit contributions | |
ittens | -0.051 |
ls | -0.049 |
shrimp | -0.041 |
strawberries | -0.040 |
she | -0.040 |
Token | rightful |
Token ID | 43133 |
Feature activation | +9.23e+00 |
Pos logit contributions | |
tted | +0.086 |
deserve | +0.082 |
construct | +0.081 |
rebuild | +0.077 |
respect | +0.077 |
Neg logit contributions | |
ittens | -0.066 |
ls | -0.063 |
ogging | -0.054 |
ule | -0.053 |
ip | -0.053 |
Token | owner |
Token ID | 4870 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
construct | +0.259 |
tted | +0.259 |
rebuild | +0.257 |
achieve | +0.239 |
deserve | +0.238 |
Neg logit contributions | |
ittens | -0.221 |
ls | -0.214 |
strawberries | -0.205 |
- | -0.199 |
shrimp | -0.199 |
Token | . |
Token ID | 13 |
Feature activation | +3.62e-01 |
Pos logit contributions | |
tted | +0.068 |
deserve | +0.065 |
construct | +0.064 |
rebuild | +0.062 |
ws | +0.062 |
Neg logit contributions | |
ittens | -0.038 |
ls | -0.036 |
shrimp | -0.029 |
ogging | -0.029 |
ip | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | +3.33e+00 |
Pos logit contributions | |
tted | +0.011 |
construct | +0.011 |
deserve | +0.011 |
rebuild | +0.010 |
ws | +0.010 |
Neg logit contributions | |
ittens | -0.008 |
ls | -0.008 |
she | -0.007 |
ogging | -0.007 |
shrimp | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
tted | +0.102 |
deserve | +0.098 |
construct | +0.096 |
rebuild | +0.092 |
ws | +0.092 |
Neg logit contributions | |
ittens | -0.082 |
ls | -0.079 |
ogging | -0.067 |
she | -0.066 |
ip | -0.066 |
Token | In |
Token ID | 818 |
Feature activation | +3.21e+00 |
Pos logit contributions | |
tted | +0.085 |
ws | +0.083 |
construct | +0.083 |
deserve | +0.081 |
rebuild | +0.079 |
Neg logit contributions | |
ittens | -0.062 |
ls | -0.061 |
she | -0.056 |
ogging | -0.054 |
shrimp | -0.052 |
Token | he |
Token ID | 339 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
tted | +0.063 |
deserve | +0.061 |
construct | +0.060 |
rebuild | +0.059 |
ws | +0.058 |
Neg logit contributions | |
ittens | -0.024 |
ls | -0.022 |
ip | -0.016 |
ogging | -0.016 |
ule | -0.016 |
Token | got |
Token ID | 1392 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
tted | +0.039 |
construct | +0.037 |
deserve | +0.036 |
rebuild | +0.035 |
ws | +0.035 |
Neg logit contributions | |
ls | -0.032 |
ittens | -0.032 |
she | -0.027 |
ip | -0.027 |
shrimp | -0.027 |
Token | it |
Token ID | 340 |
Feature activation | +2.62e+00 |
Pos logit contributions | |
tted | +0.000 |
deserve | +0.000 |
construct | +0.000 |
rebuild | +0.000 |
respect | +0.000 |
Neg logit contributions | |
ittens | -0.000 |
ls | -0.000 |
ule | -0.000 |
ogging | -0.000 |
ip | -0.000 |
Token | so |
Token ID | 523 |
Feature activation | -7.07e+00 |
Pos logit contributions | |
tted | +0.086 |
deserve | +0.082 |
construct | +0.082 |
rebuild | +0.078 |
ws | +0.078 |
Neg logit contributions | |
ittens | -0.057 |
ls | -0.055 |
shrimp | -0.046 |
she | -0.046 |
ip | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | +3.75e+00 |
Pos logit contributions | |
ittens | +0.148 |
ls | +0.128 |
ule | +0.114 |
ogging | +0.111 |
ip | +0.110 |
Neg logit contributions | |
tted | -0.251 |
deserve | -0.239 |
construct | -0.227 |
respect | -0.227 |
rebuild | -0.227 |
Token | rightful |
Token ID | 43133 |
Feature activation | +8.89e+00 |
Pos logit contributions | |
tted | +0.124 |
deserve | +0.117 |
construct | +0.117 |
rebuild | +0.112 |
ws | +0.111 |
Neg logit contributions | |
ittens | -0.083 |
ls | -0.080 |
shrimp | -0.068 |
strawberries | -0.067 |
ip | -0.066 |
Token | owner |
Token ID | 4870 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
construct | +0.290 |
tted | +0.286 |
rebuild | +0.284 |
deserve | +0.268 |
achieve | +0.259 |
Neg logit contributions | |
ls | -0.179 |
ittens | -0.171 |
strawberries | -0.159 |
strawberry | -0.158 |
- | -0.156 |
Token | could |
Token ID | 714 |
Feature activation | +2.86e+00 |
Pos logit contributions | |
tted | +0.059 |
construct | +0.058 |
deserve | +0.056 |
ws | +0.056 |
rebuild | +0.053 |
Neg logit contributions | |
ls | -0.042 |
ittens | -0.042 |
she | -0.039 |
ipop | -0.036 |
he | -0.035 |
Token | find |
Token ID | 1064 |
Feature activation | +4.31e+00 |
Pos logit contributions | |
tted | +0.058 |
construct | +0.051 |
deserve | +0.050 |
ws | +0.050 |
gen | +0.046 |
Neg logit contributions | |
ittens | -0.098 |
ls | -0.095 |
she | -0.090 |
shrimp | -0.087 |
sc | -0.086 |
Token | it |
Token ID | 340 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
tted | +0.158 |
deserve | +0.148 |
construct | +0.148 |
rebuild | +0.142 |
ws | +0.141 |
Neg logit contributions | |
ittens | -0.080 |
ls | -0.079 |
shrimp | -0.064 |
she | -0.063 |
strawberries | -0.062 |
Token | . |
Token ID | 13 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
tted | +0.085 |
construct | +0.081 |
deserve | +0.081 |
ws | +0.077 |
rebuild | +0.077 |
Neg logit contributions | |
ittens | -0.042 |
ls | -0.042 |
she | -0.034 |
shrimp | -0.033 |
ip | -0.032 |
Token | and |
Token ID | 290 |
Feature activation | -7.85e+00 |
Pos logit contributions | |
ittens | +0.023 |
ls | +0.021 |
she | +0.017 |
ip | +0.016 |
ogging | +0.016 |
Neg logit contributions | |
tted | -0.066 |
deserve | -0.065 |
construct | -0.063 |
ws | -0.062 |
respect | -0.061 |
Token | found |
Token ID | 1043 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
ittens | +0.208 |
ls | +0.205 |
she | +0.186 |
shrimp | +0.178 |
ip | +0.177 |
Neg logit contributions | |
tted | -0.218 |
deserve | -0.207 |
construct | -0.198 |
ws | -0.196 |
respect | -0.190 |
Token | the |
Token ID | 262 |
Feature activation | +3.55e+00 |
Pos logit contributions | |
tted | +0.088 |
deserve | +0.084 |
construct | +0.084 |
rebuild | +0.079 |
ws | +0.079 |
Neg logit contributions | |
ls | -0.054 |
ittens | -0.048 |
she | -0.048 |
strawberries | -0.045 |
shrimp | -0.044 |
Token | coin |
Token ID | 10752 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
tted | +0.122 |
construct | +0.116 |
rebuild | +0.112 |
ws | +0.112 |
deserve | +0.112 |
Neg logit contributions | |
ittens | -0.071 |
ls | -0.069 |
strawberries | -0.062 |
shrimp | -0.061 |
she | -0.060 |
Token | 's |
Token ID | 338 |
Feature activation | +6.84e-01 |
Pos logit contributions | |
ittens | +0.047 |
ls | +0.043 |
ogging | +0.037 |
ule | +0.037 |
shrimp | +0.036 |
Neg logit contributions | |
tted | -0.069 |
deserve | -0.065 |
construct | -0.065 |
rebuild | -0.063 |
ws | -0.061 |
Token | rightful |
Token ID | 43133 |
Feature activation | +8.49e+00 |
Pos logit contributions | |
tted | +0.021 |
deserve | +0.020 |
construct | +0.020 |
rebuild | +0.019 |
ws | +0.019 |
Neg logit contributions | |
ittens | -0.016 |
ls | -0.016 |
strawberries | -0.013 |
shrimp | -0.013 |
ip | -0.013 |
Token | place |
Token ID | 1295 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
construct | +0.268 |
tted | +0.266 |
rebuild | +0.262 |
deserve | +0.251 |
ws | +0.241 |
Neg logit contributions | |
ittens | -0.179 |
ls | -0.172 |
strawberries | -0.161 |
strawberry | -0.153 |
ipop | -0.151 |
Token | . |
Token ID | 13 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
ittens | +0.027 |
ls | +0.026 |
shrimp | +0.021 |
ogging | +0.021 |
ip | +0.021 |
Neg logit contributions | |
tted | -0.044 |
deserve | -0.042 |
construct | -0.042 |
rebuild | -0.040 |
ws | -0.040 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
ittens | +0.005 |
ls | +0.005 |
ule | +0.004 |
ogging | +0.004 |
shrimp | +0.004 |
Neg logit contributions | |
tted | -0.007 |
deserve | -0.007 |
construct | -0.007 |
rebuild | -0.006 |
ws | -0.006 |
Token | was |
Token ID | 373 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
tted | +0.061 |
deserve | +0.058 |
construct | +0.057 |
rebuild | +0.055 |
ws | +0.054 |
Neg logit contributions | |
ittens | -0.056 |
ls | -0.054 |
shrimp | -0.046 |
ogging | -0.046 |
ule | -0.046 |
Token | happy |
Token ID | 3772 |
Feature activation | -3.80e+00 |
Pos logit contributions | |
ittens | +0.021 |
ls | +0.019 |
she | +0.016 |
hungry | +0.015 |
shrimp | +0.015 |
Neg logit contributions | |
tted | -0.055 |
deserve | -0.053 |
construct | -0.053 |
ws | -0.052 |
rebuild | -0.050 |
Token | sign |
Token ID | 1051 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
tted | +0.113 |
ws | +0.087 |
deserve | +0.083 |
reed | +0.083 |
wn | +0.082 |
Neg logit contributions | |
sc | -0.127 |
ittens | -0.127 |
she | -0.122 |
because | -0.121 |
, | -0.118 |
Token | ' |
Token ID | 705 |
Feature activation | +3.09e+00 |
Pos logit contributions | |
tted | +0.042 |
deserve | +0.040 |
construct | +0.039 |
rebuild | +0.038 |
ws | +0.037 |
Neg logit contributions | |
ittens | -0.029 |
ls | -0.029 |
shrimp | -0.024 |
she | -0.024 |
ip | -0.023 |
Token | happy |
Token ID | 34191 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
tted | +0.073 |
construct | +0.070 |
deserve | +0.068 |
rebuild | +0.066 |
ws | +0.066 |
Neg logit contributions | |
ittens | -0.094 |
ls | -0.090 |
ip | -0.082 |
ries | -0.080 |
shrimp | -0.080 |
Token | ', |
Token ID | 3256 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
ittens | +0.056 |
ls | +0.054 |
shrimp | +0.049 |
she | +0.048 |
ogging | +0.048 |
Neg logit contributions | |
tted | -0.048 |
construct | -0.044 |
deserve | -0.043 |
ws | -0.042 |
rebuild | -0.041 |
Token | ' |
Token ID | 705 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
ittens | +0.034 |
ls | +0.032 |
ule | +0.026 |
ogging | +0.025 |
ip | +0.025 |
Neg logit contributions | |
tted | -0.069 |
deserve | -0.068 |
construct | -0.066 |
respect | -0.065 |
rebuild | -0.064 |
Token | birth |
Token ID | 24280 |
Feature activation | +8.42e+00 |
Pos logit contributions | |
tted | +0.029 |
construct | +0.027 |
deserve | +0.026 |
rebuild | +0.025 |
ws | +0.025 |
Neg logit contributions | |
ittens | -0.036 |
ls | -0.034 |
ip | -0.031 |
shrimp | -0.031 |
strawberries | -0.031 |
Token | day |
Token ID | 820 |
Feature activation | -5.20e+00 |
Pos logit contributions | |
tted | +0.213 |
construct | +0.199 |
hips | +0.196 |
rebuild | +0.194 |
deserve | +0.193 |
Neg logit contributions | |
shrimp | -0.217 |
ls | -0.209 |
ittens | -0.207 |
- | -0.202 |
strawberry | -0.201 |
Token | ', |
Token ID | 3256 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
ittens | +0.120 |
ls | +0.115 |
shrimp | +0.103 |
she | +0.102 |
ip | +0.099 |
Neg logit contributions | |
tted | -0.166 |
construct | -0.154 |
deserve | -0.151 |
ws | -0.149 |
rebuild | -0.147 |
Token | and |
Token ID | 290 |
Feature activation | -7.84e+00 |
Pos logit contributions | |
ittens | +0.037 |
ls | +0.034 |
shrimp | +0.026 |
ogging | +0.026 |
ip | +0.026 |
Neg logit contributions | |
tted | -0.081 |
deserve | -0.078 |
construct | -0.077 |
rebuild | -0.075 |
ws | -0.074 |
Token | ' |
Token ID | 705 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
ittens | +0.166 |
ls | +0.151 |
ule | +0.125 |
ogging | +0.124 |
ip | +0.121 |
Neg logit contributions | |
tted | -0.272 |
deserve | -0.264 |
construct | -0.257 |
rebuild | -0.250 |
respect | -0.249 |
Token | you |
Token ID | 5832 |
Feature activation | +3.90e+00 |
Pos logit contributions | |
tted | +0.040 |
construct | +0.037 |
deserve | +0.037 |
rebuild | +0.035 |
ws | +0.034 |
Neg logit contributions | |
ittens | -0.051 |
ls | -0.049 |
ip | -0.045 |
shrimp | -0.044 |
strawberries | -0.044 |
Token | wealthy |
Token ID | 11574 |
Feature activation | -3.66e-01 |
Pos logit contributions | |
ittens | +0.181 |
ls | +0.170 |
ogging | +0.140 |
ule | +0.137 |
ip | +0.137 |
Neg logit contributions | |
tted | -0.285 |
deserve | -0.275 |
construct | -0.268 |
rebuild | -0.259 |
respect | -0.256 |
Token | farmer |
Token ID | 18739 |
Feature activation | +8.98e-01 |
Pos logit contributions | |
ittens | +0.009 |
ls | +0.008 |
ogging | +0.007 |
ip | +0.007 |
ipop | +0.007 |
Neg logit contributions | |
tted | -0.011 |
construct | -0.011 |
deserve | -0.011 |
respect | -0.010 |
rebuild | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +6.29e-01 |
Pos logit contributions | |
tted | +0.032 |
deserve | +0.031 |
construct | +0.030 |
rebuild | +0.029 |
ws | +0.029 |
Neg logit contributions | |
ittens | -0.018 |
ls | -0.017 |
shrimp | -0.013 |
ip | -0.013 |
ogging | -0.013 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +6.91e+00 |
Pos logit contributions | |
tted | +0.018 |
deserve | +0.017 |
construct | +0.017 |
rebuild | +0.016 |
ws | +0.016 |
Neg logit contributions | |
ittens | -0.017 |
ls | -0.016 |
shrimp | -0.014 |
ogging | -0.014 |
ip | -0.014 |
Token | once |
Token ID | 27078 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
construct | +0.224 |
tted | +0.224 |
deserve | +0.219 |
rebuild | +0.218 |
ws | +0.216 |
Neg logit contributions | |
ls | -0.145 |
ittens | -0.139 |
she | -0.123 |
ip | -0.117 |
ogging | -0.116 |
Token | upon |
Token ID | 2402 |
Feature activation | +8.42e+00 |
Pos logit contributions | |
ittens | +0.027 |
ls | +0.025 |
shrimp | +0.022 |
ogging | +0.022 |
ip | +0.021 |
Neg logit contributions | |
tted | -0.031 |
deserve | -0.029 |
construct | -0.029 |
rebuild | -0.028 |
ws | -0.028 |
Token | time |
Token ID | 640 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
tted | +0.286 |
construct | +0.270 |
ws | +0.260 |
rebuild | +0.260 |
deserve | +0.258 |
Neg logit contributions | |
ittens | -0.170 |
ls | -0.162 |
she | -0.161 |
shrimp | -0.151 |
hungry | -0.149 |
Token | there |
Token ID | 612 |
Feature activation | +6.04e+00 |
Pos logit contributions | |
tted | +0.005 |
will | +0.005 |
deserve | +0.005 |
overcome | +0.005 |
construct | +0.005 |
Neg logit contributions | |
ittens | -0.003 |
ls | -0.002 |
strawberries | -0.002 |
ries | -0.002 |
shrimp | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
tted | +0.177 |
deserve | +0.167 |
construct | +0.165 |
rebuild | +0.158 |
ws | +0.157 |
Neg logit contributions | |
ittens | -0.157 |
ls | -0.152 |
shrimp | -0.129 |
ip | -0.129 |
she | -0.129 |
Token | a |
Token ID | 257 |
Feature activation | +8.47e-01 |
Pos logit contributions | |
tted | +0.005 |
deserve | +0.005 |
construct | +0.004 |
respect | +0.004 |
overcome | +0.004 |
Neg logit contributions | |
ittens | -0.008 |
ls | -0.008 |
ipop | -0.007 |
ogging | -0.007 |
ule | -0.007 |
Token | young |
Token ID | 1862 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
tted | +0.021 |
deserve | +0.021 |
make | +0.021 |
will | +0.021 |
construct | +0.020 |
Neg logit contributions | |
ittens | -0.023 |
ls | -0.023 |
ries | -0.023 |
ogging | -0.021 |
uts | -0.020 |
Token | should |
Token ID | 815 |
Feature activation | +4.23e+00 |
Pos logit contributions | |
tted | +0.076 |
construct | +0.074 |
ws | +0.064 |
izons | +0.061 |
directions | +0.061 |
Neg logit contributions | |
ls | -0.159 |
ittens | -0.154 |
sc | -0.143 |
she | -0.142 |
shrimp | -0.140 |
Token | be |
Token ID | 307 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
tted | +0.111 |
construct | +0.099 |
deserve | +0.092 |
ws | +0.088 |
rebuild | +0.087 |
Neg logit contributions | |
ittens | -0.129 |
ls | -0.120 |
she | -0.114 |
shrimp | -0.110 |
sc | -0.108 |
Token | compassionate |
Token ID | 32533 |
Feature activation | +9.05e-01 |
Pos logit contributions | |
tted | +0.042 |
construct | +0.041 |
ws | +0.037 |
hips | +0.037 |
deserve | +0.037 |
Neg logit contributions | |
ittens | -0.022 |
ls | -0.021 |
sc | -0.020 |
hungry | -0.020 |
she | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +6.26e-01 |
Pos logit contributions | |
tted | +0.034 |
construct | +0.033 |
deserve | +0.031 |
rebuild | +0.031 |
ws | +0.031 |
Neg logit contributions | |
ittens | -0.016 |
ls | -0.015 |
shrimp | -0.013 |
she | -0.012 |
ogging | -0.012 |
Token | Comp |
Token ID | 3082 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
tted | +0.015 |
deserve | +0.015 |
construct | +0.014 |
rebuild | +0.013 |
ws | +0.013 |
Neg logit contributions | |
ittens | -0.019 |
ls | -0.019 |
ogging | -0.016 |
shrimp | -0.016 |
ip | -0.016 |
Token | assion |
Token ID | 11857 |
Feature activation | +8.42e+00 |
Pos logit contributions | |
ls | +0.073 |
ittens | +0.070 |
ip | +0.066 |
ries | +0.061 |
ule | +0.060 |
Neg logit contributions | |
tted | -0.061 |
construct | -0.059 |
deserve | -0.055 |
overcome | -0.054 |
build | -0.053 |
Token | ate |
Token ID | 378 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
tted | +0.202 |
construct | +0.191 |
wn | +0.172 |
rebuild | +0.172 |
reed | +0.171 |
Neg logit contributions | |
ls | -0.270 |
- | -0.229 |
ittens | -0.229 |
shrimp | -0.218 |
ip | -0.212 |
Token | means |
Token ID | 1724 |
Feature activation | -2.67e-01 |
Pos logit contributions | |
tted | +0.055 |
construct | +0.050 |
deserve | +0.049 |
rebuild | +0.047 |
ws | +0.046 |
Neg logit contributions | |
ittens | -0.081 |
ls | -0.081 |
shrimp | -0.071 |
ip | -0.070 |
she | -0.070 |
Token | you |
Token ID | 345 |
Feature activation | +2.96e+00 |
Pos logit contributions | |
ittens | +0.007 |
ls | +0.007 |
ule | +0.006 |
ogging | +0.006 |
ip | +0.006 |
Neg logit contributions | |
deserve | -0.008 |
respect | -0.007 |
tted | -0.007 |
make | -0.007 |
construct | -0.007 |
Token | care |
Token ID | 1337 |
Feature activation | -8.75e-01 |
Pos logit contributions | |
tted | +0.061 |
construct | +0.057 |
rebuild | +0.050 |
ws | +0.049 |
deserve | +0.049 |
Neg logit contributions | |
ls | -0.101 |
ittens | -0.101 |
she | -0.091 |
shrimp | -0.090 |
ip | -0.090 |
Token | about |
Token ID | 546 |
Feature activation | +3.41e+00 |
Pos logit contributions | |
ittens | +0.015 |
ls | +0.015 |
she | +0.012 |
because | +0.012 |
ogging | +0.011 |
Neg logit contributions | |
tted | -0.034 |
construct | -0.032 |
deserve | -0.030 |
rebuild | -0.030 |
ws | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | +3.73e+00 |
Pos logit contributions | |
tted | +0.177 |
deserve | +0.169 |
construct | +0.166 |
rebuild | +0.160 |
ws | +0.158 |
Neg logit contributions | |
ittens | -0.149 |
ls | -0.143 |
ogging | -0.121 |
ip | -0.120 |
ule | -0.120 |
Token | valuable |
Token ID | 8119 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
tted | +0.103 |
deserve | +0.097 |
construct | +0.095 |
rebuild | +0.091 |
respect | +0.090 |
Neg logit contributions | |
ittens | -0.104 |
ls | -0.100 |
ip | -0.086 |
ogging | -0.086 |
ule | -0.085 |
Token | things |
Token ID | 1243 |
Feature activation | +3.57e-01 |
Pos logit contributions | |
tted | +0.008 |
deserve | +0.007 |
construct | +0.007 |
respect | +0.007 |
rebuild | +0.007 |
Neg logit contributions | |
ittens | -0.010 |
ls | -0.009 |
ip | -0.008 |
ogging | -0.008 |
ule | -0.008 |
Token | you |
Token ID | 345 |
Feature activation | +3.49e+00 |
Pos logit contributions | |
tted | +0.013 |
deserve | +0.013 |
construct | +0.012 |
rebuild | +0.012 |
ws | +0.012 |
Neg logit contributions | |
ittens | -0.007 |
ls | -0.006 |
shrimp | -0.005 |
ogging | -0.005 |
ip | -0.005 |
Token | already |
Token ID | 1541 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
tted | +0.084 |
construct | +0.077 |
rebuild | +0.072 |
ws | +0.069 |
deserve | +0.069 |
Neg logit contributions | |
ls | -0.103 |
ittens | -0.099 |
she | -0.096 |
shrimp | -0.095 |
hungry | -0.093 |
Token | have |
Token ID | 423 |
Feature activation | +8.25e+00 |
Pos logit contributions | |
tted | +0.068 |
construct | +0.062 |
rebuild | +0.058 |
ws | +0.058 |
deserve | +0.056 |
Neg logit contributions | |
ls | -0.071 |
she | -0.070 |
ittens | -0.069 |
because | -0.065 |
he | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
tted | +0.291 |
construct | +0.273 |
deserve | +0.273 |
rebuild | +0.262 |
ws | +0.261 |
Neg logit contributions | |
ittens | -0.164 |
ls | -0.159 |
shrimp | -0.132 |
she | -0.129 |
strawberries | -0.128 |
Token | â |
Token ID | 22940 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
tted | +0.033 |
deserve | +0.031 |
construct | +0.030 |
rebuild | +0.029 |
ws | +0.029 |
Neg logit contributions | |
ittens | -0.031 |
ls | -0.030 |
shrimp | -0.026 |
ogging | -0.026 |
ip | -0.026 |
Token | € |
Token ID | 26391 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
ittens | +0.065 |
ogging | +0.059 |
ip | +0.058 |
ls | +0.058 |
ule | +0.057 |
Neg logit contributions | |
construct | -0.049 |
tted | -0.048 |
rebuild | -0.046 |
ws | -0.044 |
respect | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | +3.78e+00 |
Pos logit contributions | |
ittens | +0.054 |
ls | +0.051 |
ip | +0.044 |
ule | +0.044 |
ogging | +0.043 |
Neg logit contributions | |
tted | -0.060 |
construct | -0.057 |
deserve | -0.057 |
ws | -0.055 |
rebuild | -0.055 |
Token |
|
Token ID | 198 |
Feature activation | +2.94e+00 |
Pos logit contributions | |
tted | +0.114 |
deserve | +0.109 |
construct | +0.109 |
rebuild | +0.104 |
ws | +0.103 |
Neg logit contributions | |
ittens | -0.095 |
ls | -0.091 |
ogging | -0.077 |
ip | -0.077 |
ule | -0.077 |
Token | The |
Token ID | 464 |
Feature activation | +3.75e+00 |
Pos logit contributions | |
tted | +0.076 |
deserve | +0.072 |
construct | +0.071 |
rebuild | +0.068 |
respect | +0.068 |
Neg logit contributions | |
ittens | -0.054 |
ls | -0.051 |
ule | -0.045 |
ogging | -0.045 |
ip | -0.043 |
Token | End |
Token ID | 5268 |
Feature activation | +9.07e-01 |
Pos logit contributions | |
tted | +0.118 |
deserve | +0.111 |
construct | +0.109 |
rebuild | +0.105 |
ws | +0.105 |
Neg logit contributions | |
ittens | -0.090 |
ls | -0.087 |
shrimp | -0.074 |
strawberries | -0.073 |
ule | -0.072 |
Token | . |
Token ID | 13 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
tted | +0.030 |
deserve | +0.029 |
construct | +0.029 |
rebuild | +0.027 |
ws | +0.027 |
Neg logit contributions | |
ittens | -0.020 |
ls | -0.019 |
shrimp | -0.016 |
ogging | -0.016 |
ip | -0.016 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +6.88e+00 |
Pos logit contributions | |
tted | +0.084 |
construct | +0.079 |
deserve | +0.077 |
ws | +0.076 |
rebuild | +0.076 |
Neg logit contributions | |
ittens | -0.061 |
ls | -0.058 |
she | -0.055 |
ogging | -0.052 |
ip | -0.051 |
Token | Once |
Token ID | 7454 |
Feature activation | -2.61e+00 |
Pos logit contributions | |
tted | +0.214 |
construct | +0.209 |
deserve | +0.207 |
rebuild | +0.202 |
ws | +0.201 |
Neg logit contributions | |
ls | -0.155 |
ittens | -0.153 |
she | -0.130 |
ogging | -0.130 |
ip | -0.129 |
Token | upon |
Token ID | 2402 |
Feature activation | +8.21e+00 |
Pos logit contributions | |
ittens | +0.074 |
ls | +0.069 |
shrimp | +0.063 |
strawberries | +0.063 |
ule | +0.061 |
Neg logit contributions | |
tted | -0.068 |
deserve | -0.063 |
construct | -0.063 |
rebuild | -0.060 |
ws | -0.060 |
Token | a |
Token ID | 257 |
Feature activation | +3.10e+00 |
Pos logit contributions | |
tted | +0.252 |
construct | +0.237 |
deserve | +0.233 |
rebuild | +0.226 |
ws | +0.226 |
Neg logit contributions | |
ittens | -0.200 |
ls | -0.191 |
she | -0.173 |
shrimp | -0.169 |
strawberries | -0.165 |
Token | time |
Token ID | 640 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
tted | +0.077 |
deserve | +0.072 |
construct | +0.070 |
rebuild | +0.067 |
ws | +0.066 |
Neg logit contributions | |
ittens | -0.095 |
ls | -0.092 |
shrimp | -0.081 |
ogging | -0.080 |
ule | -0.080 |
Token | there |
Token ID | 612 |
Feature activation | +5.53e+00 |
Pos logit contributions | |
ittens | +0.026 |
ls | +0.023 |
strawberries | +0.023 |
actus | +0.022 |
shrimp | +0.021 |
Neg logit contributions | |
tted | -0.031 |
are | -0.030 |
overcome | -0.030 |
build | -0.029 |
will | -0.029 |
Token | was |
Token ID | 373 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
tted | +0.170 |
deserve | +0.163 |
construct | +0.159 |
ws | +0.155 |
rebuild | +0.152 |
Neg logit contributions | |
ittens | -0.135 |
ls | -0.130 |
she | -0.114 |
shrimp | -0.110 |
ule | -0.110 |
Token | a |
Token ID | 257 |
Feature activation | +7.60e-01 |
Pos logit contributions | |
tted | +0.008 |
deserve | +0.007 |
overcome | +0.007 |
respect | +0.007 |
construct | +0.007 |
Neg logit contributions | |
ls | -0.013 |
ittens | -0.012 |
ipop | -0.011 |
ries | -0.011 |
ule | -0.011 |
Token | The |
Token ID | 383 |
Feature activation | +4.20e+00 |
Pos logit contributions | |
tted | +0.013 |
deserve | +0.013 |
construct | +0.012 |
respect | +0.011 |
rebuild | +0.011 |
Neg logit contributions | |
ls | -0.017 |
ittens | -0.017 |
strawberries | -0.014 |
ries | -0.014 |
ip | -0.014 |
Token | End |
Token ID | 5268 |
Feature activation | +8.11e-01 |
Pos logit contributions | |
tted | +0.128 |
deserve | +0.123 |
construct | +0.121 |
respect | +0.117 |
rebuild | +0.116 |
Neg logit contributions | |
ittens | -0.101 |
ls | -0.096 |
ogging | -0.083 |
ip | -0.081 |
ule | -0.079 |
Token | . |
Token ID | 13 |
Feature activation | +3.10e+00 |
Pos logit contributions | |
tted | +0.026 |
deserve | +0.025 |
construct | +0.024 |
respect | +0.023 |
rebuild | +0.023 |
Neg logit contributions | |
ittens | -0.019 |
ls | -0.018 |
strawberries | -0.015 |
shrimp | -0.015 |
ip | -0.015 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +7.22e+00 |
Pos logit contributions | |
tted | +0.092 |
construct | +0.086 |
deserve | +0.086 |
rebuild | +0.082 |
ws | +0.082 |
Neg logit contributions | |
ittens | -0.079 |
ls | -0.076 |
ogging | -0.066 |
she | -0.066 |
shrimp | -0.065 |
Token | Once |
Token ID | 7454 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
tted | +0.229 |
construct | +0.226 |
deserve | +0.223 |
rebuild | +0.219 |
ws | +0.217 |
Neg logit contributions | |
ls | -0.158 |
ittens | -0.155 |
ogging | -0.130 |
she | -0.130 |
ip | -0.130 |
Token | upon |
Token ID | 2402 |
Feature activation | +8.21e+00 |
Pos logit contributions | |
ittens | +0.069 |
ls | +0.064 |
strawberries | +0.058 |
shrimp | +0.058 |
ogging | +0.057 |
Neg logit contributions | |
tted | -0.064 |
deserve | -0.060 |
construct | -0.059 |
rebuild | -0.057 |
ws | -0.057 |
Token | a |
Token ID | 257 |
Feature activation | +3.03e+00 |
Pos logit contributions | |
tted | +0.255 |
construct | +0.241 |
deserve | +0.235 |
rebuild | +0.231 |
ws | +0.229 |
Neg logit contributions | |
ittens | -0.196 |
ls | -0.187 |
she | -0.170 |
shrimp | -0.166 |
strawberries | -0.161 |
Token | time |
Token ID | 640 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
tted | +0.076 |
deserve | +0.071 |
construct | +0.070 |
rebuild | +0.066 |
ws | +0.066 |
Neg logit contributions | |
ittens | -0.091 |
ls | -0.089 |
shrimp | -0.077 |
ogging | -0.077 |
ip | -0.077 |
Token | there |
Token ID | 612 |
Feature activation | +5.39e+00 |
Pos logit contributions | |
ittens | +0.031 |
ls | +0.028 |
strawberries | +0.028 |
actus | +0.026 |
ries | +0.026 |
Neg logit contributions | |
tted | -0.039 |
are | -0.037 |
overcome | -0.036 |
build | -0.036 |
am | -0.035 |
Token | was |
Token ID | 373 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
tted | +0.168 |
deserve | +0.161 |
construct | +0.157 |
ws | +0.153 |
rebuild | +0.150 |
Neg logit contributions | |
ittens | -0.130 |
ls | -0.126 |
she | -0.110 |
ule | -0.105 |
ip | -0.105 |
Token | a |
Token ID | 257 |
Feature activation | +6.38e-01 |
Pos logit contributions | |
tted | +0.002 |
deserve | +0.002 |
overcome | +0.002 |
respect | +0.002 |
construct | +0.002 |
Neg logit contributions | |
ls | -0.004 |
ittens | -0.004 |
ipop | -0.003 |
ries | -0.003 |
ule | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +2.57e+00 |
Pos logit contributions | |
tted | +0.096 |
deserve | +0.091 |
construct | +0.091 |
rebuild | +0.087 |
ws | +0.086 |
Neg logit contributions | |
ittens | -0.080 |
ls | -0.076 |
ogging | -0.065 |
ip | -0.065 |
ule | -0.065 |
Token | The |
Token ID | 464 |
Feature activation | +3.15e+00 |
Pos logit contributions | |
tted | +0.060 |
deserve | +0.056 |
construct | +0.055 |
rebuild | +0.052 |
ws | +0.052 |
Neg logit contributions | |
ittens | -0.083 |
ls | -0.079 |
ogging | -0.070 |
shrimp | -0.070 |
ip | -0.070 |
Token | End |
Token ID | 5268 |
Feature activation | +4.22e-01 |
Pos logit contributions | |
tted | +0.090 |
deserve | +0.087 |
construct | +0.084 |
respect | +0.084 |
rebuild | +0.081 |
Neg logit contributions | |
ittens | -0.081 |
ls | -0.077 |
ogging | -0.067 |
ip | -0.066 |
ries | -0.066 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +6.35e+00 |
Pos logit contributions | |
tted | +0.012 |
deserve | +0.012 |
construct | +0.011 |
respect | +0.010 |
make | +0.010 |
Neg logit contributions | |
ittens | -0.011 |
ls | -0.011 |
strawberries | -0.009 |
ries | -0.009 |
Gig | -0.009 |
Token | Once |
Token ID | 7454 |
Feature activation | -2.88e+00 |
Pos logit contributions | |
construct | +0.203 |
tted | +0.202 |
rebuild | +0.200 |
ws | +0.196 |
deserve | +0.195 |
Neg logit contributions | |
ls | -0.132 |
ittens | -0.130 |
ogging | -0.113 |
she | -0.111 |
shrimp | -0.110 |
Token | upon |
Token ID | 2402 |
Feature activation | +8.19e+00 |
Pos logit contributions | |
ittens | +0.080 |
ls | +0.074 |
shrimp | +0.067 |
strawberries | +0.066 |
ule | +0.065 |
Neg logit contributions | |
tted | -0.078 |
deserve | -0.073 |
construct | -0.073 |
rebuild | -0.070 |
ws | -0.069 |
Token | a |
Token ID | 257 |
Feature activation | +3.04e+00 |
Pos logit contributions | |
tted | +0.246 |
construct | +0.230 |
deserve | +0.227 |
rebuild | +0.220 |
ws | +0.219 |
Neg logit contributions | |
ittens | -0.206 |
ls | -0.198 |
she | -0.175 |
shrimp | -0.174 |
strawberries | -0.170 |
Token | time |
Token ID | 640 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
tted | +0.072 |
deserve | +0.068 |
construct | +0.065 |
rebuild | +0.062 |
respect | +0.062 |
Neg logit contributions | |
ittens | -0.095 |
ls | -0.093 |
shrimp | -0.082 |
ries | -0.081 |
ule | -0.081 |
Token | there |
Token ID | 612 |
Feature activation | +5.51e+00 |
Pos logit contributions | |
ittens | +0.028 |
ls | +0.025 |
strawberries | +0.025 |
actus | +0.025 |
ries | +0.023 |
Neg logit contributions | |
tted | -0.031 |
are | -0.030 |
overcome | -0.029 |
build | -0.029 |
am | -0.029 |
Token | was |
Token ID | 373 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
tted | +0.158 |
deserve | +0.149 |
construct | +0.147 |
rebuild | +0.141 |
ws | +0.139 |
Neg logit contributions | |
ittens | -0.147 |
ls | -0.142 |
shrimp | -0.121 |
ogging | -0.121 |
ip | -0.121 |
Token | a |
Token ID | 257 |
Feature activation | +6.17e-01 |
Pos logit contributions | |
tted | +0.003 |
deserve | +0.003 |
respect | +0.003 |
gen | +0.003 |
overcome | +0.003 |
Neg logit contributions | |
ls | -0.005 |
ittens | -0.005 |
ipop | -0.004 |
ule | -0.004 |
ries | -0.004 |
Token | bouncing |
Token ID | 30713 |
Feature activation | +4.38e-01 |
Pos logit contributions | |
ittens | +0.090 |
ls | +0.084 |
ip | +0.066 |
ule | +0.066 |
ries | +0.064 |
Neg logit contributions | |
tted | -0.167 |
deserve | -0.160 |
construct | -0.156 |
rebuild | -0.154 |
overcome | -0.151 |
Token | around |
Token ID | 1088 |
Feature activation | +3.43e+00 |
Pos logit contributions | |
tted | +0.011 |
make | +0.011 |
build | +0.010 |
construct | +0.010 |
respect | +0.010 |
Neg logit contributions | |
ittens | -0.013 |
ule | -0.011 |
ls | -0.011 |
ries | -0.011 |
ip | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | -9.33e+00 |
Pos logit contributions | |
construct | +0.100 |
make | +0.098 |
build | +0.096 |
tted | +0.096 |
rebuild | +0.096 |
Neg logit contributions | |
ittens | -0.087 |
ls | -0.084 |
ipop | -0.075 |
ip | -0.073 |
ogging | -0.069 |
Token | an |
Token ID | 281 |
Feature activation | -8.08e-01 |
Pos logit contributions | |
ittens | +0.166 |
ls | +0.159 |
ries | +0.142 |
ipop | +0.141 |
ip | +0.130 |
Neg logit contributions | |
overcome | -0.326 |
make | -0.325 |
build | -0.320 |
tted | -0.320 |
construct | -0.317 |
Token | unpleasant |
Token ID | 22029 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
ls | +0.021 |
ittens | +0.020 |
ipop | +0.019 |
ip | +0.019 |
ogging | +0.017 |
Neg logit contributions | |
tted | -0.022 |
construct | -0.021 |
deserve | -0.021 |
rebuild | -0.021 |
build | -0.020 |
Token | od |
Token ID | 16298 |
Feature activation | -1.54e+01 |
Pos logit contributions | |
ittens | +0.048 |
ls | +0.046 |
ries | +0.038 |
ip | +0.038 |
ipop | +0.038 |
Neg logit contributions | |
tted | -0.070 |
deserve | -0.068 |
construct | -0.066 |
make | -0.064 |
respect | -0.063 |
Token | our |
Token ID | 454 |
Feature activation | +8.72e-01 |
Pos logit contributions | |
ittens | +0.349 |
she | +0.341 |
he | +0.323 |
peanuts | +0.319 |
Gig | +0.314 |
Neg logit contributions | |
tted | -0.429 |
construct | -0.404 |
ws | -0.395 |
rebuild | -0.380 |
hips | -0.378 |
Token | wa |
Token ID | 2082 |
Feature activation | -9.41e+00 |
Pos logit contributions | |
tted | +0.026 |
deserve | +0.025 |
construct | +0.024 |
ws | +0.024 |
rebuild | +0.024 |
Neg logit contributions | |
ittens | -0.023 |
ls | -0.022 |
ule | -0.019 |
she | -0.019 |
ip | -0.018 |
Token | ft |
Token ID | 701 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
ittens | +0.247 |
ls | +0.241 |
ip | +0.205 |
ule | +0.203 |
ogging | +0.202 |
Neg logit contributions | |
tted | -0.271 |
deserve | -0.260 |
construct | -0.258 |
ws | -0.247 |
rebuild | -0.246 |
Token | ed |
Token ID | 276 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
tted | +0.039 |
construct | +0.038 |
deserve | +0.038 |
rebuild | +0.037 |
ws | +0.036 |
Neg logit contributions | |
ls | -0.025 |
ittens | -0.025 |
ip | -0.020 |
ogging | -0.020 |
she | -0.020 |
Token | into |
Token ID | 656 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
ittens | +0.051 |
ls | +0.045 |
ule | +0.041 |
ogging | +0.041 |
ip | +0.041 |
Neg logit contributions | |
tted | -0.053 |
construct | -0.050 |
deserve | -0.049 |
rebuild | -0.048 |
ws | -0.048 |
Token | suddenly |
Token ID | 6451 |
Feature activation | -3.74e+00 |
Pos logit contributions | |
ittens | +0.072 |
ls | +0.052 |
ogging | +0.051 |
ip | +0.048 |
ule | +0.046 |
Neg logit contributions | |
overcome | -0.134 |
make | -0.133 |
will | -0.128 |
construct | -0.128 |
create | -0.127 |
Token | , |
Token ID | 11 |
Feature activation | -9.56e+00 |
Pos logit contributions | |
ittens | +0.092 |
ogging | +0.076 |
ule | +0.071 |
ls | +0.070 |
ip | +0.064 |
Neg logit contributions | |
make | -0.110 |
construct | -0.102 |
overcome | -0.102 |
deserve | -0.102 |
form | -0.101 |
Token | a |
Token ID | 257 |
Feature activation | -3.20e-01 |
Pos logit contributions | |
ittens | +0.190 |
ls | +0.146 |
ogging | +0.140 |
ule | +0.129 |
ip | +0.121 |
Neg logit contributions | |
deserve | -0.332 |
make | -0.331 |
tted | -0.328 |
construct | -0.324 |
overcome | -0.322 |
Token | brightly |
Token ID | 35254 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
ittens | +0.009 |
ls | +0.008 |
ries | +0.008 |
ip | +0.007 |
ogging | +0.007 |
Neg logit contributions | |
tted | -0.009 |
deserve | -0.008 |
construct | -0.008 |
make | -0.008 |
form | -0.008 |
Token | - |
Token ID | 12 |
Feature activation | -9.02e+00 |
Pos logit contributions | |
ittens | +0.042 |
ls | +0.039 |
ip | +0.037 |
ogging | +0.035 |
strawberries | +0.035 |
Neg logit contributions | |
tted | -0.036 |
deserve | -0.034 |
construct | -0.034 |
make | -0.033 |
form | -0.033 |
Token | col |
Token ID | 4033 |
Feature activation | -1.44e+01 |
Pos logit contributions | |
ittens | +0.224 |
ls | +0.211 |
ogging | +0.179 |
shrimp | +0.178 |
strawberries | +0.177 |
Neg logit contributions | |
tted | -0.286 |
deserve | -0.265 |
construct | -0.255 |
respect | -0.249 |
ws | -0.245 |
Token | oured |
Token ID | 8167 |
Feature activation | -3.34e+00 |
Pos logit contributions | |
ittens | +0.415 |
ls | +0.407 |
she | +0.387 |
shrimp | +0.371 |
hungry | +0.361 |
Neg logit contributions | |
tted | -0.355 |
ws | -0.330 |
construct | -0.327 |
deserve | -0.316 |
rebuild | -0.311 |
Token | belt |
Token ID | 10999 |
Feature activation | -4.02e+00 |
Pos logit contributions | |
ittens | +0.086 |
ls | +0.084 |
ogging | +0.072 |
ip | +0.071 |
ries | +0.068 |
Neg logit contributions | |
tted | -0.093 |
construct | -0.088 |
form | -0.087 |
deserve | -0.086 |
make | -0.085 |
Token | caught |
Token ID | 4978 |
Feature activation | -4.86e-01 |
Pos logit contributions | |
ittens | +0.095 |
ls | +0.089 |
shrimp | +0.076 |
ip | +0.075 |
ogging | +0.075 |
Neg logit contributions | |
tted | -0.129 |
deserve | -0.121 |
construct | -0.120 |
rebuild | -0.115 |
ws | -0.113 |
Token | her |
Token ID | 607 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
ittens | +0.013 |
ls | +0.011 |
ogging | +0.011 |
ule | +0.011 |
ipop | +0.010 |
Neg logit contributions | |
tted | -0.013 |
deserve | -0.012 |
construct | -0.012 |
ws | -0.012 |
respect | -0.012 |
Token | eye |
Token ID | 4151 |
Feature activation | -2.53e+00 |
Pos logit contributions | |
tted | +0.026 |
deserve | +0.024 |
construct | +0.024 |
rebuild | +0.022 |
ws | +0.022 |
Neg logit contributions | |
ittens | -0.031 |
ls | -0.030 |
shrimp | -0.026 |
ip | -0.026 |
ogging | -0.026 |
Token | She |
Token ID | 1375 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
tted | +0.008 |
deserve | +0.007 |
construct | +0.007 |
respect | +0.007 |
ws | +0.006 |
Neg logit contributions | |
ittens | -0.008 |
ls | -0.008 |
shrimp | -0.007 |
ip | -0.007 |
ule | -0.007 |
Token | also |
Token ID | 635 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
tted | +0.059 |
construct | +0.056 |
deserve | +0.055 |
make | +0.053 |
ws | +0.053 |
Neg logit contributions | |
ittens | -0.069 |
ls | -0.063 |
ogging | -0.059 |
ule | -0.057 |
peanuts | -0.056 |
Token | put |
Token ID | 1234 |
Feature activation | +6.93e-01 |
Pos logit contributions | |
tted | +0.001 |
deserve | +0.001 |
construct | +0.001 |
rebuild | +0.001 |
ws | +0.001 |
Neg logit contributions | |
ittens | -0.001 |
ls | -0.001 |
shrimp | -0.000 |
ip | -0.000 |
she | -0.000 |
Token | some |
Token ID | 617 |
Feature activation | +4.18e+00 |
Pos logit contributions | |
tted | +0.019 |
deserve | +0.018 |
ws | +0.018 |
make | +0.018 |
construct | +0.017 |
Neg logit contributions | |
ittens | -0.018 |
ls | -0.017 |
ule | -0.017 |
ogging | -0.016 |
yt | -0.015 |
Token | lip |
Token ID | 10645 |
Feature activation | -5.00e+00 |
Pos logit contributions | |
tted | +0.126 |
deserve | +0.116 |
respect | +0.115 |
construct | +0.115 |
make | +0.114 |
Neg logit contributions | |
ittens | -0.101 |
ls | -0.095 |
ule | -0.090 |
ogging | -0.085 |
Gig | -0.084 |
Token | bal |
Token ID | 3652 |
Feature activation | -1.34e+01 |
Pos logit contributions | |
ls | +0.115 |
Nem | +0.107 |
ittens | +0.107 |
Gig | +0.106 |
Twe | +0.103 |
Neg logit contributions | |
tted | -0.150 |
age | -0.140 |
make | -0.128 |
construct | -0.126 |
respect | -0.125 |
Token | m |
Token ID | 76 |
Feature activation | -3.92e+00 |
Pos logit contributions | |
ittens | +0.310 |
ls | +0.302 |
shrimp | +0.261 |
ule | +0.253 |
ogging | +0.251 |
Neg logit contributions | |
tted | -0.424 |
deserve | -0.396 |
construct | -0.389 |
ws | -0.377 |
rebuild | -0.375 |
Token | on |
Token ID | 319 |
Feature activation | +3.39e+00 |
Pos logit contributions | |
ls | +0.086 |
ittens | +0.082 |
ogging | +0.072 |
Gig | +0.069 |
shrimp | +0.068 |
Neg logit contributions | |
tted | -0.124 |
deserve | -0.116 |
ws | -0.116 |
construct | -0.114 |
respect | -0.113 |
Token | her |
Token ID | 607 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
make | +0.086 |
construct | +0.083 |
form | +0.083 |
respect | +0.081 |
tted | +0.081 |
Neg logit contributions | |
ls | -0.099 |
ittens | -0.095 |
ogging | -0.083 |
ule | -0.083 |
ipop | -0.083 |
Token | lips |
Token ID | 11914 |
Feature activation | +8.95e-01 |
Pos logit contributions | |
tted | +0.052 |
construct | +0.048 |
make | +0.047 |
deserve | +0.047 |
respect | +0.046 |
Neg logit contributions | |
ls | -0.060 |
ittens | -0.056 |
ogging | -0.052 |
Gig | -0.050 |
ule | -0.049 |
Token | to |
Token ID | 284 |
Feature activation | +3.35e+00 |
Pos logit contributions | |
tted | +0.031 |
construct | +0.029 |
deserve | +0.029 |
make | +0.028 |
respect | +0.028 |
Neg logit contributions | |
ls | -0.018 |
ittens | -0.017 |
shrimp | -0.013 |
ule | -0.013 |
ogging | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | -8.49e-02 |
Pos logit contributions | |
respect | +0.011 |
make | +0.010 |
overcome | +0.010 |
construct | +0.010 |
rebuild | +0.010 |
Neg logit contributions | |
ittens | -0.021 |
ls | -0.019 |
ipop | -0.018 |
ip | -0.018 |
ule | -0.017 |
Token | strange |
Token ID | 6283 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
ittens | +0.002 |
ls | +0.002 |
ries | +0.002 |
ogging | +0.002 |
ip | +0.002 |
Neg logit contributions | |
form | -0.002 |
tted | -0.002 |
deserve | -0.002 |
make | -0.002 |
will | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -1.11e+01 |
Pos logit contributions | |
ittens | +0.047 |
ls | +0.041 |
ip | +0.039 |
ries | +0.036 |
ogging | +0.036 |
Neg logit contributions | |
form | -0.035 |
make | -0.033 |
construct | -0.033 |
deserve | -0.032 |
tted | -0.031 |
Token | raspberry |
Token ID | 38973 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
ittens | +0.322 |
ls | +0.272 |
ip | +0.265 |
ule | +0.255 |
ipop | +0.251 |
Neg logit contributions | |
deserve | -0.308 |
make | -0.307 |
tted | -0.303 |
form | -0.301 |
exact | -0.294 |
Token | - |
Token ID | 12 |
Feature activation | -7.39e+00 |
Pos logit contributions | |
ittens | +0.054 |
ls | +0.049 |
ogging | +0.049 |
Nem | +0.046 |
Twe | +0.045 |
Neg logit contributions | |
tted | -0.061 |
make | -0.060 |
construct | -0.058 |
deserve | -0.056 |
age | -0.056 |
Token | col |
Token ID | 4033 |
Feature activation | -1.34e+01 |
Pos logit contributions | |
ittens | +0.211 |
ls | +0.197 |
ogging | +0.183 |
yt | +0.170 |
shrimp | +0.169 |
Neg logit contributions | |
tted | -0.198 |
deserve | -0.188 |
construct | -0.178 |
ws | -0.177 |
gen | -0.176 |
Token | oured |
Token ID | 8167 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
ittens | +0.379 |
ls | +0.375 |
she | +0.365 |
he | +0.346 |
ule | +0.342 |
Neg logit contributions | |
ws | -0.319 |
tted | -0.319 |
construct | -0.306 |
deserve | -0.291 |
rebuild | -0.287 |
Token | object |
Token ID | 2134 |
Feature activation | -2.90e+00 |
Pos logit contributions | |
ittens | +0.066 |
ls | +0.062 |
ogging | +0.054 |
ip | +0.050 |
ule | +0.048 |
Neg logit contributions | |
tted | -0.091 |
deserve | -0.090 |
make | -0.090 |
construct | -0.089 |
form | -0.088 |
Token | hiding |
Token ID | 11816 |
Feature activation | +6.58e-01 |
Pos logit contributions | |
ittens | +0.082 |
ls | +0.069 |
Gig | +0.067 |
ogging | +0.067 |
Nem | +0.065 |
Neg logit contributions | |
tted | -0.071 |
make | -0.069 |
construct | -0.067 |
will | -0.064 |
build | -0.064 |
Token | under |
Token ID | 739 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
tted | +0.023 |
construct | +0.022 |
deserve | +0.021 |
rebuild | +0.021 |
ws | +0.021 |
Neg logit contributions | |
ittens | -0.014 |
ls | -0.012 |
ogging | -0.011 |
ule | -0.011 |
ip | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | +9.10e-01 |
Pos logit contributions | |
tted | +0.065 |
deserve | +0.062 |
construct | +0.062 |
respect | +0.059 |
rebuild | +0.059 |
Neg logit contributions | |
ittens | -0.071 |
ls | -0.067 |
ogging | -0.060 |
ule | -0.060 |
ip | -0.059 |
Token | day |
Token ID | 1110 |
Feature activation | -3.96e+00 |
Pos logit contributions | |
ittens | +0.067 |
ls | +0.061 |
ogging | +0.054 |
ries | +0.053 |
shrimp | +0.052 |
Neg logit contributions | |
tted | -0.085 |
deserve | -0.078 |
construct | -0.075 |
gen | -0.073 |
will | -0.072 |
Token | , |
Token ID | 11 |
Feature activation | -8.92e+00 |
Pos logit contributions | |
lee | +0.101 |
ittens | +0.098 |
ries | +0.093 |
ogging | +0.092 |
ls | +0.092 |
Neg logit contributions | |
are | -0.092 |
make | -0.090 |
tted | -0.089 |
am | -0.089 |
will | -0.086 |
Token | a |
Token ID | 257 |
Feature activation | +6.81e-02 |
Pos logit contributions | |
ogging | +0.227 |
ittens | +0.226 |
ls | +0.221 |
lee | +0.219 |
ule | +0.215 |
Neg logit contributions | |
make | -0.234 |
tted | -0.227 |
will | -0.221 |
deserve | -0.220 |
are | -0.220 |
Token | chocolate |
Token ID | 11311 |
Feature activation | -4.90e+00 |
Pos logit contributions | |
tted | +0.002 |
deserve | +0.002 |
make | +0.002 |
will | +0.002 |
construct | +0.002 |
Neg logit contributions | |
ls | -0.002 |
ittens | -0.002 |
ries | -0.002 |
ogging | -0.001 |
ip | -0.001 |
Token | - |
Token ID | 12 |
Feature activation | -6.99e+00 |
Pos logit contributions | |
ittens | +0.121 |
ogging | +0.118 |
ls | +0.117 |
lee | +0.111 |
Nem | +0.102 |
Neg logit contributions | |
make | -0.128 |
will | -0.127 |
age | -0.126 |
tted | -0.125 |
deserve | -0.121 |
Token | col |
Token ID | 4033 |
Feature activation | -1.32e+01 |
Pos logit contributions | |
ittens | +0.155 |
ls | +0.147 |
ogging | +0.125 |
shrimp | +0.124 |
strawberries | +0.119 |
Neg logit contributions | |
tted | -0.234 |
deserve | -0.221 |
construct | -0.215 |
ws | -0.208 |
rebuild | -0.206 |
Token | oured |
Token ID | 8167 |
Feature activation | -2.99e+00 |
Pos logit contributions | |
ls | +0.376 |
ittens | +0.372 |
she | +0.364 |
hungry | +0.348 |
shrimp | +0.347 |
Neg logit contributions | |
tted | -0.306 |
ws | -0.300 |
construct | -0.277 |
deserve | -0.268 |
respect | -0.266 |
Token | whip |
Token ID | 22972 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
ls | +0.069 |
ittens | +0.065 |
ogging | +0.058 |
lee | +0.054 |
ip | +0.053 |
Neg logit contributions | |
tted | -0.086 |
make | -0.085 |
deserve | -0.084 |
respect | -0.083 |
will | -0.083 |
Token | named |
Token ID | 3706 |
Feature activation | -4.51e+00 |
Pos logit contributions | |
ittens | +0.052 |
ls | +0.048 |
ogging | +0.041 |
strawberries | +0.040 |
ip | +0.040 |
Neg logit contributions | |
tted | -0.063 |
deserve | -0.059 |
construct | -0.059 |
make | -0.057 |
rebuild | -0.056 |
Token | C |
Token ID | 327 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
ittens | +0.086 |
ls | +0.082 |
ogging | +0.066 |
shrimp | +0.064 |
ule | +0.063 |
Neg logit contributions | |
tted | -0.163 |
deserve | -0.157 |
construct | -0.154 |
rebuild | -0.148 |
ws | -0.148 |
Token | heddar |
Token ID | 44937 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
ittens | +0.033 |
ls | +0.031 |
ip | +0.027 |
ule | +0.026 |
ogging | +0.026 |
Neg logit contributions | |
tted | -0.045 |
deserve | -0.043 |
construct | -0.043 |
rebuild | -0.041 |
ws | -0.040 |
Token | In |
Token ID | 554 |
Feature activation | +3.58e+00 |
Pos logit contributions | |
ittens | +0.003 |
ls | +0.003 |
shrimp | +0.002 |
strawberries | +0.002 |
ogging | +0.002 |
Neg logit contributions | |
tted | -0.004 |
deserve | -0.004 |
construct | -0.004 |
rebuild | -0.003 |
ws | -0.003 |
Token | fact |
Token ID | 1109 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
tted | +0.115 |
deserve | +0.109 |
construct | +0.108 |
rebuild | +0.104 |
ws | +0.103 |
Neg logit contributions | |
ittens | -0.084 |
ls | -0.080 |
shrimp | -0.067 |
ip | -0.067 |
ogging | -0.067 |
Token | , |
Token ID | 11 |
Feature activation | -9.32e+00 |
Pos logit contributions | |
ittens | +0.040 |
ls | +0.032 |
ule | +0.029 |
strawberries | +0.029 |
shrimp | +0.027 |
Neg logit contributions | |
tted | -0.099 |
deserve | -0.095 |
construct | -0.093 |
overcome | -0.092 |
rebuild | -0.092 |
Token | the |
Token ID | 262 |
Feature activation | +3.64e+00 |
Pos logit contributions | |
ittens | +0.169 |
ls | +0.160 |
ogging | +0.127 |
ule | +0.127 |
ipop | +0.123 |
Neg logit contributions | |
tted | -0.339 |
deserve | -0.331 |
construct | -0.329 |
make | -0.320 |
rebuild | -0.320 |
Token | special |
Token ID | 2041 |
Feature activation | -7.23e-02 |
Pos logit contributions | |
tted | +0.110 |
deserve | +0.104 |
construct | +0.102 |
rebuild | +0.098 |
ws | +0.097 |
Neg logit contributions | |
ittens | -0.093 |
ls | -0.089 |
shrimp | -0.075 |
ogging | -0.075 |
ip | -0.075 |
Token | bal |
Token ID | 3652 |
Feature activation | -1.31e+01 |
Pos logit contributions | |
ittens | +0.002 |
ls | +0.002 |
shrimp | +0.002 |
ogging | +0.002 |
ip | +0.002 |
Neg logit contributions | |
tted | -0.002 |
deserve | -0.002 |
construct | -0.002 |
rebuild | -0.002 |
ws | -0.002 |
Token | m |
Token ID | 76 |
Feature activation | -3.63e+00 |
Pos logit contributions | |
ittens | +0.330 |
ls | +0.317 |
ip | +0.270 |
ogging | +0.269 |
ule | +0.266 |
Neg logit contributions | |
tted | -0.397 |
deserve | -0.377 |
construct | -0.372 |
rebuild | -0.357 |
ws | -0.354 |
Token | made |
Token ID | 925 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
ittens | +0.075 |
ls | +0.073 |
shrimp | +0.062 |
strawberries | +0.061 |
Gig | +0.060 |
Neg logit contributions | |
tted | -0.121 |
deserve | -0.113 |
construct | -0.112 |
rebuild | -0.107 |
respect | -0.107 |
Token | her |
Token ID | 607 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
tted | +0.115 |
deserve | +0.111 |
construct | +0.109 |
ws | +0.108 |
rebuild | +0.104 |
Neg logit contributions | |
ittens | -0.041 |
ls | -0.039 |
she | -0.035 |
shrimp | -0.029 |
ogging | -0.028 |
Token | ankle |
Token ID | 19847 |
Feature activation | -4.56e+00 |
Pos logit contributions | |
tted | +0.034 |
deserve | +0.031 |
construct | +0.030 |
ws | +0.028 |
rebuild | +0.028 |
Neg logit contributions | |
ittens | -0.063 |
ls | -0.061 |
ogging | -0.055 |
ip | -0.055 |
she | -0.055 |
Token | look |
Token ID | 804 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
ittens | +0.131 |
ls | +0.129 |
she | +0.124 |
ogging | +0.115 |
sc | +0.115 |
Neg logit contributions | |
tted | -0.113 |
deserve | -0.106 |
ws | -0.105 |
construct | -0.105 |
gen | -0.103 |
Token | the |
Token ID | 262 |
Feature activation | +3.84e+00 |
Pos logit contributions | |
tted | +0.116 |
construct | +0.110 |
deserve | +0.109 |
rebuild | +0.106 |
respect | +0.102 |
Neg logit contributions | |
ittens | -0.106 |
ls | -0.102 |
ip | -0.087 |
ule | -0.087 |
ogging | -0.086 |
Token | store |
Token ID | 3650 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
tted | +0.120 |
deserve | +0.114 |
construct | +0.112 |
rebuild | +0.108 |
ws | +0.107 |
Neg logit contributions | |
ittens | -0.093 |
ls | -0.089 |
shrimp | -0.075 |
ogging | -0.074 |
ip | -0.074 |
Token | and |
Token ID | 290 |
Feature activation | -7.32e+00 |
Pos logit contributions | |
ittens | +0.046 |
ls | +0.043 |
shrimp | +0.032 |
ip | +0.032 |
ule | +0.032 |
Neg logit contributions | |
tted | -0.109 |
construct | -0.103 |
deserve | -0.103 |
rebuild | -0.101 |
overcome | -0.098 |
Token | got |
Token ID | 1392 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ittens | +0.237 |
ls | +0.222 |
ule | +0.202 |
ogging | +0.196 |
ries | +0.190 |
Neg logit contributions | |
tted | -0.181 |
construct | -0.165 |
deserve | -0.163 |
make | -0.161 |
rebuild | -0.161 |
Token | the |
Token ID | 262 |
Feature activation | +3.47e+00 |
Pos logit contributions | |
ittens | +0.035 |
ls | +0.033 |
ule | +0.030 |
ogging | +0.029 |
ip | +0.028 |
Neg logit contributions | |
tted | -0.035 |
deserve | -0.032 |
construct | -0.032 |
rebuild | -0.031 |
respect | -0.031 |
Token | bal |
Token ID | 3652 |
Feature activation | -1.31e+01 |
Pos logit contributions | |
tted | +0.111 |
deserve | +0.105 |
construct | +0.104 |
rebuild | +0.101 |
ws | +0.100 |
Neg logit contributions | |
ittens | -0.083 |
ls | -0.078 |
strawberries | -0.067 |
she | -0.066 |
ogging | -0.066 |
Token | m |
Token ID | 76 |
Feature activation | -4.20e+00 |
Pos logit contributions | |
ittens | +0.293 |
ls | +0.280 |
ip | +0.232 |
ogging | +0.232 |
ule | +0.228 |
Neg logit contributions | |
tted | -0.433 |
deserve | -0.413 |
construct | -0.408 |
rebuild | -0.393 |
ws | -0.390 |
Token | . |
Token ID | 13 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
ittens | +0.090 |
ls | +0.090 |
Gig | +0.077 |
shrimp | +0.077 |
strawberries | +0.074 |
Neg logit contributions | |
tted | -0.133 |
deserve | -0.123 |
construct | -0.121 |
ws | -0.117 |
respect | -0.117 |
Token | She |
Token ID | 1375 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
ittens | +0.001 |
ls | +0.001 |
ule | +0.000 |
strawberries | +0.000 |
ries | +0.000 |
Neg logit contributions | |
tted | -0.001 |
deserve | -0.001 |
construct | -0.001 |
rebuild | -0.001 |
overcome | -0.001 |
Token | carefully |
Token ID | 7773 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
tted | +0.052 |
construct | +0.046 |
deserve | +0.046 |
make | +0.045 |
overcome | +0.044 |
Neg logit contributions | |
ittens | -0.060 |
ls | -0.059 |
ogging | -0.054 |
ule | -0.052 |
shrimp | -0.051 |
Token | rubbed |
Token ID | 31862 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
ittens | +0.041 |
ls | +0.040 |
ule | +0.036 |
ogging | +0.034 |
shrimp | +0.034 |
Neg logit contributions | |
tted | -0.038 |
construct | -0.035 |
deserve | -0.034 |
rebuild | -0.033 |
ws | -0.032 |
Token | date |
Token ID | 3128 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
tted | +0.141 |
deserve | +0.132 |
construct | +0.131 |
rebuild | +0.127 |
ws | +0.126 |
Neg logit contributions | |
ittens | -0.099 |
ls | -0.097 |
strawberries | -0.082 |
shrimp | -0.081 |
she | -0.080 |
Token | , |
Token ID | 11 |
Feature activation | -9.03e+00 |
Pos logit contributions | |
tted | +0.095 |
deserve | +0.094 |
construct | +0.091 |
ws | +0.090 |
rebuild | +0.089 |
Neg logit contributions | |
ittens | -0.028 |
ls | -0.028 |
she | -0.024 |
ip | -0.020 |
ipop | -0.019 |
Token | a |
Token ID | 257 |
Feature activation | +3.52e-01 |
Pos logit contributions | |
ittens | +0.156 |
ls | +0.135 |
ogging | +0.113 |
ip | +0.111 |
ule | +0.111 |
Neg logit contributions | |
tted | -0.345 |
deserve | -0.340 |
construct | -0.333 |
make | -0.325 |
rebuild | -0.324 |
Token | rainbow |
Token ID | 27223 |
Feature activation | -2.41e+00 |
Pos logit contributions | |
tted | +0.011 |
deserve | +0.011 |
construct | +0.011 |
rebuild | +0.010 |
overcome | +0.010 |
Neg logit contributions | |
ittens | -0.008 |
ls | -0.008 |
ogging | -0.007 |
ip | -0.006 |
ries | -0.006 |
Token | - |
Token ID | 12 |
Feature activation | -6.79e+00 |
Pos logit contributions | |
ittens | +0.061 |
ls | +0.059 |
shrimp | +0.050 |
ip | +0.050 |
ogging | +0.050 |
Neg logit contributions | |
tted | -0.072 |
deserve | -0.068 |
construct | -0.067 |
rebuild | -0.065 |
ws | -0.065 |
Token | col |
Token ID | 4033 |
Feature activation | -1.30e+01 |
Pos logit contributions | |
ittens | +0.184 |
ls | +0.180 |
ip | +0.154 |
shrimp | +0.153 |
ogging | +0.152 |
Neg logit contributions | |
tted | -0.190 |
deserve | -0.179 |
construct | -0.178 |
rebuild | -0.170 |
ws | -0.168 |
Token | oured |
Token ID | 8167 |
Feature activation | -2.63e+00 |
Pos logit contributions | |
ittens | +0.371 |
ls | +0.359 |
she | +0.331 |
shrimp | +0.329 |
ule | +0.328 |
Neg logit contributions | |
tted | -0.339 |
construct | -0.311 |
ws | -0.310 |
deserve | -0.308 |
rebuild | -0.295 |
Token | butterfly |
Token ID | 35113 |
Feature activation | -9.88e-01 |
Pos logit contributions | |
ittens | +0.070 |
ls | +0.064 |
ogging | +0.056 |
ip | +0.056 |
ule | +0.055 |
Neg logit contributions | |
tted | -0.076 |
deserve | -0.073 |
construct | -0.073 |
rebuild | -0.069 |
respect | -0.069 |
Token | flew |
Token ID | 13112 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
ittens | +0.024 |
ls | +0.021 |
ule | +0.019 |
ogging | +0.019 |
shrimp | +0.019 |
Neg logit contributions | |
tted | -0.031 |
deserve | -0.030 |
construct | -0.029 |
rebuild | -0.028 |
respect | -0.027 |
Token | in |
Token ID | 287 |
Feature activation | +3.77e+00 |
Pos logit contributions | |
ittens | +0.046 |
ls | +0.044 |
shrimp | +0.038 |
ogging | +0.038 |
ip | +0.038 |
Neg logit contributions | |
tted | -0.048 |
deserve | -0.045 |
construct | -0.045 |
rebuild | -0.043 |
ws | -0.042 |
Token | through |
Token ID | 832 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
tted | +0.129 |
deserve | +0.121 |
construct | +0.119 |
ws | +0.116 |
rebuild | +0.115 |
Neg logit contributions | |
ittens | -0.078 |
ls | -0.077 |
she | -0.065 |
strawberries | -0.065 |
shrimp | -0.064 |
Token | ribbon |
Token ID | 29092 |
Feature activation | -7.08e-01 |
Pos logit contributions | |
ittens | +0.023 |
ls | +0.022 |
ip | +0.019 |
ule | +0.019 |
ogging | +0.019 |
Neg logit contributions | |
tted | -0.024 |
deserve | -0.022 |
construct | -0.022 |
respect | -0.021 |
rebuild | -0.021 |
Token | for |
Token ID | 329 |
Feature activation | +5.50e+00 |
Pos logit contributions | |
ittens | +0.012 |
ls | +0.012 |
shrimp | +0.009 |
ogging | +0.009 |
ule | +0.009 |
Neg logit contributions | |
tted | -0.027 |
deserve | -0.026 |
construct | -0.025 |
rebuild | -0.025 |
ws | -0.024 |
Token | being |
Token ID | 852 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
tted | +0.175 |
deserve | +0.168 |
construct | +0.168 |
rebuild | +0.161 |
respect | +0.161 |
Neg logit contributions | |
ittens | -0.125 |
ls | -0.122 |
ip | -0.101 |
ule | -0.101 |
ogging | -0.100 |
Token | the |
Token ID | 262 |
Feature activation | +3.50e+00 |
Pos logit contributions | |
tted | +0.062 |
deserve | +0.059 |
construct | +0.058 |
rebuild | +0.056 |
ws | +0.056 |
Neg logit contributions | |
ittens | -0.047 |
ls | -0.045 |
shrimp | -0.039 |
ip | -0.038 |
strawberries | -0.038 |
Token | best |
Token ID | 1266 |
Feature activation | -4.85e+00 |
Pos logit contributions | |
tted | +0.096 |
construct | +0.088 |
ws | +0.086 |
deserve | +0.086 |
rebuild | +0.085 |
Neg logit contributions | |
ittens | -0.093 |
ls | -0.088 |
strawberries | -0.084 |
shrimp | -0.084 |
she | -0.083 |
Token | bal |
Token ID | 3652 |
Feature activation | -1.30e+01 |
Pos logit contributions | |
ittens | +0.140 |
ls | +0.135 |
shrimp | +0.116 |
ip | +0.116 |
ule | +0.116 |
Neg logit contributions | |
tted | -0.130 |
deserve | -0.123 |
construct | -0.120 |
rebuild | -0.114 |
respect | -0.114 |
Token | ancer |
Token ID | 8250 |
Feature activation | +2.86e+00 |
Pos logit contributions | |
ittens | +0.345 |
ls | +0.333 |
ip | +0.291 |
ogging | +0.289 |
ipop | +0.286 |
Neg logit contributions | |
tted | -0.372 |
deserve | -0.356 |
construct | -0.354 |
rebuild | -0.338 |
ws | -0.336 |
Token | in |
Token ID | 287 |
Feature activation | +4.22e+00 |
Pos logit contributions | |
tted | +0.100 |
deserve | +0.095 |
construct | +0.094 |
rebuild | +0.091 |
ws | +0.090 |
Neg logit contributions | |
ittens | -0.058 |
ls | -0.055 |
shrimp | -0.044 |
ogging | -0.044 |
ip | -0.044 |
Token | the |
Token ID | 262 |
Feature activation | +3.92e+00 |
Pos logit contributions | |
tted | +0.136 |
deserve | +0.128 |
construct | +0.126 |
rebuild | +0.121 |
ws | +0.121 |
Neg logit contributions | |
ittens | -0.098 |
ls | -0.094 |
shrimp | -0.079 |
ogging | -0.078 |
she | -0.078 |
Token | competition |
Token ID | 5449 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
tted | +0.112 |
ws | +0.102 |
construct | +0.100 |
deserve | +0.099 |
rebuild | +0.097 |
Neg logit contributions | |
ittens | -0.097 |
ls | -0.095 |
shrimp | -0.091 |
she | -0.090 |
hungry | -0.089 |
Token | . |
Token ID | 13 |
Feature activation | +7.39e-01 |
Pos logit contributions | |
ittens | +0.026 |
ls | +0.025 |
she | +0.021 |
ogging | +0.020 |
ip | +0.020 |
Neg logit contributions | |
tted | -0.054 |
deserve | -0.051 |
construct | -0.051 |
ws | -0.049 |
rebuild | -0.049 |
Token | , |
Token ID | 11 |
Feature activation | -1.00e+01 |
Pos logit contributions | |
ittens | +0.007 |
ls | +0.006 |
ule | +0.005 |
ogging | +0.005 |
ip | +0.005 |
Neg logit contributions | |
tted | -0.010 |
deserve | -0.009 |
will | -0.009 |
make | -0.009 |
construct | -0.009 |
Token | with |
Token ID | 351 |
Feature activation | +9.05e-01 |
Pos logit contributions | |
ittens | +0.242 |
ls | +0.226 |
ule | +0.197 |
ogging | +0.190 |
ip | +0.187 |
Neg logit contributions | |
deserve | -0.318 |
make | -0.316 |
tted | -0.306 |
construct | -0.301 |
overcome | -0.297 |
Token | all |
Token ID | 477 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
tted | +0.030 |
construct | +0.027 |
deserve | +0.027 |
rebuild | +0.026 |
ws | +0.026 |
Neg logit contributions | |
ittens | -0.020 |
ls | -0.020 |
she | -0.018 |
strawberries | -0.018 |
shrimp | -0.017 |
Token | its |
Token ID | 663 |
Feature activation | +2.65e+00 |
Pos logit contributions | |
tted | +0.036 |
deserve | +0.033 |
construct | +0.032 |
rebuild | +0.031 |
ws | +0.031 |
Neg logit contributions | |
ittens | -0.023 |
ls | -0.023 |
she | -0.020 |
strawberries | -0.020 |
shrimp | -0.019 |
Token | different |
Token ID | 1180 |
Feature activation | +9.43e-01 |
Pos logit contributions | |
tted | +0.089 |
deserve | +0.082 |
construct | +0.082 |
rebuild | +0.080 |
ws | +0.079 |
Neg logit contributions | |
ittens | -0.060 |
ls | -0.057 |
shrimp | -0.048 |
strawberries | -0.047 |
she | -0.047 |
Token | col |
Token ID | 4033 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
tted | +0.030 |
deserve | +0.027 |
construct | +0.026 |
rebuild | +0.026 |
ws | +0.025 |
Neg logit contributions | |
ittens | -0.023 |
ls | -0.023 |
shrimp | -0.020 |
strawberries | -0.020 |
she | -0.019 |
Token | ours |
Token ID | 4662 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
ls | +0.319 |
ittens | +0.313 |
she | +0.305 |
he | +0.295 |
ule | +0.292 |
Neg logit contributions | |
tted | -0.352 |
ws | -0.345 |
construct | -0.332 |
deserve | -0.331 |
rebuild | -0.312 |
Token | , |
Token ID | 11 |
Feature activation | -9.93e+00 |
Pos logit contributions | |
tted | +0.068 |
deserve | +0.065 |
construct | +0.063 |
ws | +0.061 |
respect | +0.061 |
Neg logit contributions | |
ittens | -0.043 |
ls | -0.038 |
ule | -0.034 |
ogging | -0.033 |
ip | -0.032 |
Token | shapes |
Token ID | 15268 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
ittens | +0.227 |
ls | +0.211 |
ule | +0.182 |
ogging | +0.180 |
ip | +0.176 |
Neg logit contributions | |
tted | -0.324 |
deserve | -0.315 |
construct | -0.309 |
rebuild | -0.299 |
respect | -0.296 |
Token | and |
Token ID | 290 |
Feature activation | -8.37e+00 |
Pos logit contributions | |
tted | +0.043 |
deserve | +0.041 |
construct | +0.041 |
rebuild | +0.040 |
ws | +0.039 |
Neg logit contributions | |
ittens | -0.019 |
ls | -0.018 |
ule | -0.013 |
ogging | -0.013 |
ip | -0.013 |
Token | textures |
Token ID | 20028 |
Feature activation | -7.27e-01 |
Pos logit contributions | |
ittens | +0.223 |
ls | +0.218 |
shrimp | +0.190 |
strawberries | +0.190 |
she | +0.188 |
Neg logit contributions | |
tted | -0.241 |
deserve | -0.221 |
construct | -0.218 |
rebuild | -0.209 |
ws | -0.208 |
Token | . |
Token ID | 13 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
ittens | +0.050 |
ls | +0.046 |
ip | +0.039 |
shrimp | +0.038 |
ule | +0.037 |
Neg logit contributions | |
tted | -0.074 |
deserve | -0.071 |
construct | -0.070 |
rebuild | -0.067 |
respect | -0.066 |
Token | It |
Token ID | 632 |
Feature activation | +3.24e+00 |
Pos logit contributions | |
tted | +0.006 |
deserve | +0.005 |
construct | +0.005 |
make | +0.004 |
build | +0.004 |
Neg logit contributions | |
ls | -0.010 |
ittens | -0.010 |
ries | -0.009 |
ule | -0.009 |
ogging | -0.009 |
Token | was |
Token ID | 373 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
tted | +0.083 |
deserve | +0.079 |
construct | +0.077 |
make | +0.072 |
rebuild | +0.072 |
Neg logit contributions | |
ittens | -0.100 |
ls | -0.091 |
ogging | -0.085 |
ip | -0.083 |
ule | -0.083 |
Token | a |
Token ID | 257 |
Feature activation | -8.04e-01 |
Pos logit contributions | |
ittens | +0.056 |
ls | +0.055 |
ogging | +0.050 |
ip | +0.049 |
ule | +0.048 |
Neg logit contributions | |
tted | -0.039 |
construct | -0.037 |
deserve | -0.037 |
make | -0.036 |
respect | -0.035 |
Token | big |
Token ID | 1263 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
ittens | +0.024 |
ls | +0.022 |
ries | +0.021 |
ogging | +0.020 |
ip | +0.020 |
Neg logit contributions | |
deserve | -0.020 |
tted | -0.020 |
make | -0.020 |
will | -0.020 |
form | -0.020 |
Token | p |
Token ID | 279 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
ittens | +0.038 |
ls | +0.035 |
ip | +0.030 |
ries | +0.029 |
ule | +0.029 |
Neg logit contributions | |
make | -0.042 |
deserve | -0.042 |
form | -0.041 |
construct | -0.041 |
tted | -0.040 |
Token | uddle |
Token ID | 24500 |
Feature activation | -5.13e+00 |
Pos logit contributions | |
ittens | +0.255 |
strawberries | +0.239 |
Gig | +0.237 |
she | +0.228 |
Nem | +0.227 |
Neg logit contributions | |
tted | -0.431 |
deserve | -0.374 |
construct | -0.371 |
hips | -0.357 |
age | -0.356 |
Token | of |
Token ID | 286 |
Feature activation | +2.74e+00 |
Pos logit contributions | |
ittens | +0.101 |
ls | +0.093 |
ip | +0.075 |
ule | +0.074 |
strawberries | +0.074 |
Neg logit contributions | |
tted | -0.187 |
deserve | -0.176 |
construct | -0.173 |
rebuild | -0.167 |
respect | -0.164 |
Token | ink |
Token ID | 16882 |
Feature activation | -3.87e+00 |
Pos logit contributions | |
tted | +0.084 |
deserve | +0.081 |
construct | +0.080 |
rebuild | +0.077 |
respect | +0.076 |
Neg logit contributions | |
ittens | -0.068 |
ls | -0.064 |
ip | -0.056 |
ogging | -0.055 |
ule | -0.054 |
Token | ! |
Token ID | 0 |
Feature activation | +5.30e-01 |
Pos logit contributions | |
ittens | +0.086 |
ls | +0.075 |
ule | +0.066 |
Gig | +0.064 |
ogging | +0.064 |
Neg logit contributions | |
tted | -0.130 |
deserve | -0.122 |
construct | -0.121 |
make | -0.116 |
rebuild | -0.115 |
Token |
|
Token ID | 198 |
Feature activation | +3.29e+00 |
Pos logit contributions | |
tted | +0.013 |
deserve | +0.012 |
construct | +0.011 |
overcome | +0.010 |
make | +0.010 |
Neg logit contributions | |
ls | -0.017 |
ittens | -0.016 |
ule | -0.014 |
ries | -0.014 |
ogging | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | -9.33e+00 |
Pos logit contributions | |
tted | +0.175 |
construct | +0.164 |
rebuild | +0.162 |
deserve | +0.161 |
overcome | +0.161 |
Neg logit contributions | |
ittens | -0.075 |
ls | -0.061 |
ule | -0.051 |
ip | -0.048 |
strawberries | -0.048 |
Token | they |
Token ID | 484 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
ittens | +0.194 |
ule | +0.155 |
ls | +0.141 |
ogging | +0.139 |
ipop | +0.124 |
Neg logit contributions | |
tted | -0.329 |
deserve | -0.326 |
make | -0.323 |
overcome | -0.316 |
rebuild | -0.307 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
tted | +0.068 |
deserve | +0.064 |
construct | +0.063 |
rebuild | +0.060 |
ws | +0.060 |
Neg logit contributions | |
ittens | -0.064 |
ls | -0.063 |
shrimp | -0.054 |
ip | -0.053 |
ogging | -0.053 |
Token | a |
Token ID | 257 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
tted | +0.056 |
deserve | +0.055 |
respect | +0.053 |
construct | +0.053 |
rebuild | +0.053 |
Neg logit contributions | |
ittens | -0.053 |
ls | -0.047 |
ogging | -0.043 |
ule | -0.042 |
ipop | -0.042 |
Token | big |
Token ID | 1263 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
ittens | +0.004 |
ls | +0.004 |
ries | +0.003 |
ogging | +0.003 |
ip | +0.003 |
Neg logit contributions | |
tted | -0.004 |
deserve | -0.004 |
construct | -0.004 |
will | -0.004 |
form | -0.004 |
Token | oct |
Token ID | 19318 |
Feature activation | -1.27e+01 |
Pos logit contributions | |
ittens | +0.031 |
ls | +0.026 |
ip | +0.025 |
uts | +0.023 |
ries | +0.023 |
Neg logit contributions | |
make | -0.029 |
form | -0.028 |
construct | -0.027 |
deserve | -0.027 |
will | -0.027 |
Token | opus |
Token ID | 25790 |
Feature activation | -2.75e+00 |
Pos logit contributions | |
ittens | +0.189 |
ls | +0.146 |
ries | +0.145 |
strawberries | +0.138 |
Gig | +0.137 |
Neg logit contributions | |
tted | -0.519 |
deserve | -0.484 |
construct | -0.465 |
overcome | -0.452 |
gen | -0.443 |
Token | sitting |
Token ID | 5586 |
Feature activation | +6.61e-01 |
Pos logit contributions | |
ittens | +0.065 |
ls | +0.054 |
strawberries | +0.048 |
Gig | +0.048 |
ip | +0.048 |
Neg logit contributions | |
tted | -0.090 |
deserve | -0.085 |
construct | -0.085 |
rebuild | -0.083 |
make | -0.081 |
Token | on |
Token ID | 319 |
Feature activation | +3.74e+00 |
Pos logit contributions | |
tted | +0.020 |
deserve | +0.018 |
construct | +0.017 |
ws | +0.017 |
overcome | +0.017 |
Neg logit contributions | |
ittens | -0.017 |
ls | -0.015 |
ule | -0.014 |
ogging | -0.014 |
wo | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | +4.73e+00 |
Pos logit contributions | |
tted | +0.102 |
deserve | +0.096 |
construct | +0.095 |
rebuild | +0.091 |
ws | +0.089 |
Neg logit contributions | |
ittens | -0.107 |
ls | -0.101 |
ogging | -0.089 |
ule | -0.089 |
ip | -0.088 |
Token | sheet |
Token ID | 9629 |
Feature activation | -5.40e-01 |
Pos logit contributions | |
tted | +0.138 |
deserve | +0.130 |
construct | +0.128 |
rebuild | +0.123 |
ws | +0.122 |
Neg logit contributions | |
ittens | -0.124 |
ls | -0.121 |
shrimp | -0.104 |
ule | -0.102 |
strawberries | -0.102 |
Token | delicious |
Token ID | 12625 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
ittens | +0.003 |
ls | +0.003 |
ip | +0.003 |
ogging | +0.003 |
ipop | +0.003 |
Neg logit contributions | |
tted | -0.003 |
deserve | -0.003 |
construct | -0.003 |
respect | -0.003 |
make | -0.003 |
Token | vegetables |
Token ID | 13701 |
Feature activation | +5.58e-01 |
Pos logit contributions | |
ittens | +0.040 |
ipop | +0.038 |
ip | +0.035 |
ogging | +0.035 |
ule | +0.035 |
Neg logit contributions | |
make | -0.028 |
construct | -0.027 |
form | -0.027 |
deserve | -0.026 |
respect | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | -8.42e+00 |
Pos logit contributions | |
tted | +0.017 |
deserve | +0.016 |
make | +0.016 |
construct | +0.016 |
rebuild | +0.016 |
Neg logit contributions | |
ittens | -0.013 |
ls | -0.012 |
ogging | -0.011 |
ipop | -0.011 |
ule | -0.010 |
Token | bright |
Token ID | 6016 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
ittens | +0.212 |
ls | +0.195 |
ogging | +0.190 |
ule | +0.174 |
ip | +0.173 |
Neg logit contributions | |
deserve | -0.250 |
make | -0.245 |
tted | -0.245 |
construct | -0.244 |
build | -0.235 |
Token | - |
Token ID | 12 |
Feature activation | -7.62e+00 |
Pos logit contributions | |
ittens | +0.031 |
ule | +0.028 |
ogging | +0.028 |
ipop | +0.028 |
ls | +0.027 |
Neg logit contributions | |
construct | -0.023 |
make | -0.023 |
deserve | -0.023 |
respect | -0.023 |
form | -0.022 |
Token | col |
Token ID | 4033 |
Feature activation | -1.27e+01 |
Pos logit contributions | |
ittens | +0.179 |
ls | +0.169 |
ogging | +0.150 |
ipop | +0.142 |
shrimp | +0.141 |
Neg logit contributions | |
tted | -0.244 |
deserve | -0.230 |
construct | -0.219 |
gen | -0.214 |
respect | -0.213 |
Token | oured |
Token ID | 8167 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
ittens | +0.349 |
ls | +0.343 |
she | +0.332 |
hungry | +0.324 |
he | +0.323 |
Neg logit contributions | |
tted | -0.305 |
ws | -0.302 |
construct | -0.289 |
deserve | -0.283 |
rebuild | -0.272 |
Token | flowers |
Token ID | 12734 |
Feature activation | +8.05e-01 |
Pos logit contributions | |
ittens | +0.075 |
ls | +0.065 |
ogging | +0.064 |
ip | +0.062 |
ipop | +0.062 |
Neg logit contributions | |
deserve | -0.073 |
construct | -0.073 |
tted | -0.072 |
respect | -0.072 |
make | -0.072 |
Token | . |
Token ID | 13 |
Feature activation | +9.73e-01 |
Pos logit contributions | |
tted | +0.022 |
deserve | +0.021 |
construct | +0.021 |
make | +0.020 |
rebuild | +0.020 |
Neg logit contributions | |
ittens | -0.022 |
ls | -0.019 |
ogging | -0.018 |
ule | -0.018 |
ipop | -0.018 |
Token | |
Token ID | 220 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
tted | +0.021 |
deserve | +0.020 |
construct | +0.020 |
rebuild | +0.018 |
build | +0.018 |
Neg logit contributions | |
ls | -0.032 |
ittens | -0.032 |
ogging | -0.028 |
ip | -0.028 |
ule | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | +3.50e+00 |
Pos logit contributions | |
ittens | +0.030 |
ls | +0.028 |
she | +0.024 |
ip | +0.023 |
ule | +0.023 |
Neg logit contributions | |
tted | -0.047 |
construct | -0.046 |
deserve | -0.046 |
ws | -0.044 |
rebuild | -0.044 |
Token | his |
Token ID | 465 |
Feature activation | +6.53e-01 |
Pos logit contributions | |
tted | +0.095 |
deserve | +0.090 |
construct | +0.089 |
rebuild | +0.085 |
ws | +0.084 |
Neg logit contributions | |
ittens | -0.085 |
ls | -0.081 |
ogging | -0.070 |
ip | -0.069 |
ule | -0.069 |
Token | friends |
Token ID | 2460 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
tted | +0.021 |
deserve | +0.019 |
construct | +0.019 |
overcome | +0.019 |
respect | +0.018 |
Neg logit contributions | |
ittens | -0.014 |
ls | -0.014 |
ries | -0.012 |
ogging | -0.012 |
ip | -0.012 |
Token | dropped |
Token ID | 5710 |
Feature activation | -6.35e-01 |
Pos logit contributions | |
tted | +0.065 |
deserve | +0.060 |
construct | +0.059 |
overcome | +0.058 |
rebuild | +0.057 |
Neg logit contributions | |
ittens | -0.071 |
ls | -0.065 |
strawberries | -0.059 |
ipop | -0.057 |
ip | -0.057 |
Token | her |
Token ID | 607 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
ittens | +0.020 |
ls | +0.018 |
ogging | +0.017 |
ule | +0.017 |
ip | +0.017 |
Neg logit contributions | |
respect | -0.015 |
make | -0.014 |
ws | -0.014 |
tted | -0.014 |
construct | -0.014 |
Token | lip |
Token ID | 10645 |
Feature activation | -4.39e+00 |
Pos logit contributions | |
tted | +0.056 |
deserve | +0.052 |
construct | +0.051 |
respect | +0.050 |
rebuild | +0.050 |
Neg logit contributions | |
ittens | -0.046 |
ls | -0.044 |
ogging | -0.038 |
ip | -0.037 |
ule | -0.037 |
Token | bal |
Token ID | 3652 |
Feature activation | -1.27e+01 |
Pos logit contributions | |
ittens | +0.097 |
ls | +0.096 |
shrimp | +0.090 |
Gig | +0.088 |
ogging | +0.085 |
Neg logit contributions | |
tted | -0.139 |
age | -0.127 |
make | -0.121 |
deserve | -0.119 |
will | -0.116 |
Token | m |
Token ID | 76 |
Feature activation | -3.44e+00 |
Pos logit contributions | |
ittens | +0.306 |
ls | +0.299 |
shrimp | +0.267 |
strawberries | +0.260 |
Gig | +0.255 |
Neg logit contributions | |
tted | -0.382 |
deserve | -0.349 |
construct | -0.344 |
rebuild | -0.331 |
ws | -0.331 |
Token | on |
Token ID | 319 |
Feature activation | +3.42e+00 |
Pos logit contributions | |
ittens | +0.075 |
ls | +0.074 |
Gig | +0.064 |
strawberries | +0.062 |
ogging | +0.061 |
Neg logit contributions | |
tted | -0.106 |
deserve | -0.097 |
ws | -0.096 |
construct | -0.095 |
respect | -0.094 |
Token | the |
Token ID | 262 |
Feature activation | +4.25e+00 |
Pos logit contributions | |
construct | +0.075 |
make | +0.073 |
form | +0.072 |
rebuild | +0.070 |
tted | +0.070 |
Neg logit contributions | |
ls | -0.109 |
ittens | -0.109 |
ipop | -0.099 |
ogging | -0.097 |
ip | -0.097 |
Token | ground |
Token ID | 2323 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
tted | +0.103 |
deserve | +0.096 |
construct | +0.094 |
rebuild | +0.089 |
ws | +0.088 |
Neg logit contributions | |
ittens | -0.133 |
ls | -0.129 |
ogging | -0.113 |
ip | -0.113 |
shrimp | -0.112 |
Token | . |
Token ID | 13 |
Feature activation | +9.19e-01 |
Pos logit contributions | |
ittens | +0.024 |
ls | +0.022 |
shrimp | +0.018 |
ip | +0.018 |
ogging | +0.018 |
Neg logit contributions | |
tted | -0.037 |
deserve | -0.035 |
construct | -0.035 |
rebuild | -0.034 |
ws | -0.033 |
Token | The |
Token ID | 383 |
Feature activation | +3.64e+00 |
Pos logit contributions | |
tted | +0.048 |
deserve | +0.045 |
construct | +0.045 |
rebuild | +0.043 |
ws | +0.042 |
Neg logit contributions | |
ittens | -0.032 |
ls | -0.032 |
ule | -0.027 |
ogging | -0.026 |
ip | -0.026 |
Token | park |
Token ID | 3952 |
Feature activation | -3.07e+00 |
Pos logit contributions | |
tted | +0.115 |
deserve | +0.108 |
construct | +0.106 |
rebuild | +0.102 |
respect | +0.101 |
Neg logit contributions | |
ittens | -0.088 |
ls | -0.085 |
ogging | -0.072 |
ip | -0.072 |
ule | -0.070 |
Token | became |
Token ID | 2627 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
ittens | +0.093 |
ls | +0.086 |
ogging | +0.078 |
ip | +0.076 |
ule | +0.076 |
Neg logit contributions | |
tted | -0.082 |
deserve | -0.075 |
construct | -0.073 |
rebuild | -0.070 |
overcome | -0.068 |
Token | a |
Token ID | 257 |
Feature activation | -5.78e-01 |
Pos logit contributions | |
ittens | +0.040 |
ls | +0.039 |
ogging | +0.033 |
ule | +0.033 |
ipop | +0.032 |
Neg logit contributions | |
tted | -0.063 |
construct | -0.059 |
deserve | -0.059 |
rebuild | -0.057 |
overcome | -0.057 |
Token | big |
Token ID | 1263 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
ittens | +0.017 |
ls | +0.016 |
ries | +0.016 |
ogging | +0.014 |
ip | +0.014 |
Neg logit contributions | |
tted | -0.015 |
deserve | -0.015 |
will | -0.014 |
form | -0.014 |
construct | -0.014 |
Token | p |
Token ID | 279 |
Feature activation | -1.26e+01 |
Pos logit contributions | |
ittens | +0.043 |
ls | +0.040 |
ip | +0.034 |
ries | +0.034 |
ule | +0.033 |
Neg logit contributions | |
make | -0.039 |
will | -0.038 |
build | -0.038 |
deserve | -0.038 |
form | -0.038 |
Token | uddle |
Token ID | 24500 |
Feature activation | -5.08e+00 |
Pos logit contributions | |
ittens | +0.243 |
strawberries | +0.236 |
shrimp | +0.233 |
Gig | +0.231 |
Nem | +0.223 |
Neg logit contributions | |
tted | -0.429 |
deserve | -0.367 |
construct | -0.361 |
age | -0.355 |
hips | -0.349 |
Token | . |
Token ID | 13 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
ittens | +0.097 |
ls | +0.092 |
ip | +0.072 |
shrimp | +0.072 |
ogging | +0.072 |
Neg logit contributions | |
tted | -0.186 |
deserve | -0.176 |
construct | -0.174 |
rebuild | -0.169 |
ws | -0.165 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
tted | +0.037 |
deserve | +0.034 |
construct | +0.034 |
rebuild | +0.032 |
ws | +0.031 |
Neg logit contributions | |
ittens | -0.030 |
ls | -0.029 |
ule | -0.025 |
ogging | -0.024 |
ip | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | -7.57e+00 |
Pos logit contributions | |
tted | +0.079 |
deserve | +0.075 |
construct | +0.074 |
make | +0.073 |
will | +0.072 |
Neg logit contributions | |
ittens | -0.059 |
ls | -0.054 |
ries | -0.047 |
peanuts | -0.045 |
Gig | -0.045 |
Token | Ben |
Token ID | 3932 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
ittens | +0.197 |
ls | +0.193 |
ule | +0.159 |
ogging | +0.157 |
ip | +0.154 |
Neg logit contributions | |
tted | -0.201 |
construct | -0.200 |
deserve | -0.198 |
ws | -0.198 |
make | -0.197 |
Token | She |
Token ID | 1375 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
tted | +0.026 |
construct | +0.025 |
deserve | +0.025 |
ws | +0.024 |
rebuild | +0.024 |
Neg logit contributions | |
ittens | -0.023 |
ls | -0.022 |
she | -0.020 |
ip | -0.020 |
ogging | -0.019 |
Token | became |
Token ID | 2627 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
tted | +0.065 |
deserve | +0.061 |
construct | +0.060 |
rebuild | +0.057 |
ws | +0.057 |
Neg logit contributions | |
ittens | -0.065 |
ls | -0.064 |
ip | -0.056 |
shrimp | -0.055 |
she | -0.055 |
Token | known |
Token ID | 1900 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
ittens | +0.030 |
ls | +0.029 |
she | +0.025 |
shrimp | +0.025 |
strawberries | +0.025 |
Neg logit contributions | |
tted | -0.049 |
deserve | -0.046 |
construct | -0.046 |
ws | -0.045 |
rebuild | -0.044 |
Token | as |
Token ID | 355 |
Feature activation | -3.82e+00 |
Pos logit contributions | |
ittens | +0.025 |
ls | +0.022 |
ule | +0.019 |
ip | +0.018 |
ogging | +0.018 |
Neg logit contributions | |
tted | -0.046 |
deserve | -0.044 |
construct | -0.043 |
rebuild | -0.043 |
respect | -0.043 |
Token | the |
Token ID | 262 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
ittens | +0.084 |
ls | +0.078 |
ogging | +0.065 |
ip | +0.065 |
ule | +0.065 |
Neg logit contributions | |
tted | -0.128 |
deserve | -0.123 |
construct | -0.121 |
rebuild | -0.117 |
respect | -0.115 |
Token | resc |
Token ID | 6811 |
Feature activation | -1.26e+01 |
Pos logit contributions | |
tted | +0.089 |
construct | +0.087 |
ws | +0.086 |
deserve | +0.085 |
rebuild | +0.084 |
Neg logit contributions | |
ittens | -0.047 |
ls | -0.045 |
shrimp | -0.039 |
strawberries | -0.037 |
she | -0.036 |
Token | uer |
Token ID | 15573 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
ittens | +0.345 |
ls | +0.340 |
ip | +0.290 |
ogging | +0.286 |
ule | +0.285 |
Neg logit contributions | |
construct | -0.335 |
tted | -0.327 |
deserve | -0.326 |
ws | -0.322 |
respect | -0.310 |
Token | , |
Token ID | 11 |
Feature activation | -9.74e+00 |
Pos logit contributions | |
ls | +0.026 |
ittens | +0.026 |
she | +0.021 |
shrimp | +0.021 |
ogging | +0.020 |
Neg logit contributions | |
tted | -0.058 |
construct | -0.055 |
deserve | -0.055 |
ws | -0.054 |
rebuild | -0.053 |
Token | and |
Token ID | 290 |
Feature activation | -8.29e+00 |
Pos logit contributions | |
ittens | +0.189 |
ls | +0.179 |
ogging | +0.142 |
ule | +0.141 |
ip | +0.141 |
Neg logit contributions | |
tted | -0.353 |
deserve | -0.338 |
construct | -0.333 |
rebuild | -0.323 |
ws | -0.319 |
Token | was |
Token ID | 373 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
ls | +0.136 |
ittens | +0.132 |
she | +0.123 |
shrimp | +0.108 |
strawberries | +0.106 |
Neg logit contributions | |
tted | -0.318 |
deserve | -0.298 |
construct | -0.297 |
ws | -0.292 |
rebuild | -0.283 |
Token | loved |
Token ID | 6151 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
ittens | +0.039 |
ls | +0.038 |
she | +0.033 |
shrimp | +0.032 |
ip | +0.032 |
Neg logit contributions | |
tted | -0.050 |
deserve | -0.047 |
construct | -0.047 |
ws | -0.045 |
rebuild | -0.045 |
Token | acle |
Token ID | 6008 |
Feature activation | +3.38e-01 |
Pos logit contributions | |
ittens | +0.060 |
ls | +0.059 |
she | +0.048 |
shrimp | +0.047 |
ogging | +0.047 |
Neg logit contributions | |
tted | -0.105 |
deserve | -0.101 |
construct | -0.100 |
ws | -0.097 |
rebuild | -0.096 |
Token | . |
Token ID | 13 |
Feature activation | +6.42e-01 |
Pos logit contributions | |
tted | +0.011 |
deserve | +0.010 |
construct | +0.010 |
rebuild | +0.010 |
respect | +0.010 |
Neg logit contributions | |
ittens | -0.008 |
ls | -0.007 |
ip | -0.006 |
ule | -0.006 |
ogging | -0.006 |
Token | It |
Token ID | 632 |
Feature activation | +3.22e+00 |
Pos logit contributions | |
tted | +0.015 |
deserve | +0.013 |
build | +0.013 |
construct | +0.012 |
make | +0.012 |
Neg logit contributions | |
ls | -0.022 |
ittens | -0.021 |
ogging | -0.019 |
strawberries | -0.018 |
ip | -0.018 |
Token | was |
Token ID | 373 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
tted | +0.076 |
deserve | +0.071 |
construct | +0.068 |
make | +0.066 |
will | +0.064 |
Neg logit contributions | |
ittens | -0.108 |
ls | -0.096 |
ogging | -0.094 |
ip | -0.091 |
strawberries | -0.090 |
Token | an |
Token ID | 281 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
ittens | +0.055 |
ls | +0.054 |
ogging | +0.050 |
ip | +0.049 |
ipop | +0.047 |
Neg logit contributions | |
tted | -0.042 |
construct | -0.040 |
make | -0.039 |
deserve | -0.039 |
overcome | -0.037 |
Token | oct |
Token ID | 19318 |
Feature activation | -1.26e+01 |
Pos logit contributions | |
ls | +0.049 |
ittens | +0.047 |
ip | +0.044 |
ipop | +0.044 |
ries | +0.041 |
Neg logit contributions | |
tted | -0.045 |
deserve | -0.043 |
build | -0.042 |
construct | -0.041 |
belong | -0.041 |
Token | opus |
Token ID | 25790 |
Feature activation | -2.77e+00 |
Pos logit contributions | |
ittens | +0.178 |
strawberries | +0.166 |
ries | +0.162 |
ls | +0.159 |
Gig | +0.156 |
Neg logit contributions | |
tted | -0.510 |
deserve | -0.464 |
construct | -0.436 |
age | -0.429 |
overcome | -0.427 |
Token | ! |
Token ID | 0 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
ittens | +0.070 |
ls | +0.065 |
strawberries | +0.061 |
ries | +0.060 |
ipop | +0.056 |
Neg logit contributions | |
tted | -0.078 |
deserve | -0.076 |
make | -0.073 |
age | -0.072 |
will | -0.071 |
Token | The |
Token ID | 383 |
Feature activation | +3.35e+00 |
Pos logit contributions | |
tted | +0.028 |
deserve | +0.026 |
build | +0.024 |
construct | +0.024 |
make | +0.023 |
Neg logit contributions | |
ls | -0.032 |
ittens | -0.032 |
strawberries | -0.028 |
ipop | -0.028 |
ogging | -0.028 |
Token | oct |
Token ID | 19318 |
Feature activation | -1.14e+01 |
Pos logit contributions | |
tted | +0.127 |
deserve | +0.126 |
construct | +0.121 |
make | +0.119 |
overcome | +0.117 |
Neg logit contributions | |
ittens | -0.057 |
ls | -0.050 |
ries | -0.042 |
ip | -0.041 |
ogging | -0.040 |
Token | opus |
Token ID | 25790 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
ittens | +0.188 |
Gig | +0.169 |
strawberries | +0.165 |
ls | +0.151 |
ries | +0.149 |
Neg logit contributions | |
tted | -0.445 |
deserve | -0.404 |
construct | -0.385 |
overcome | -0.379 |
ars | -0.374 |
Token | a |
Token ID | 257 |
Feature activation | +6.25e-01 |
Pos logit contributions | |
tted | +0.070 |
deserve | +0.066 |
construct | +0.065 |
rebuild | +0.063 |
ws | +0.061 |
Neg logit contributions | |
ittens | -0.053 |
ls | -0.052 |
shrimp | -0.044 |
strawberries | -0.044 |
ip | -0.044 |
Token | necklace |
Token ID | 38987 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
tted | +0.020 |
deserve | +0.018 |
construct | +0.018 |
rebuild | +0.018 |
ws | +0.017 |
Neg logit contributions | |
ittens | -0.015 |
ls | -0.014 |
shrimp | -0.012 |
ip | -0.012 |
ogging | -0.012 |
Token | with |
Token ID | 351 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
ittens | +0.033 |
ls | +0.030 |
ogging | +0.024 |
shrimp | +0.024 |
ule | +0.024 |
Neg logit contributions | |
tted | -0.071 |
deserve | -0.067 |
construct | -0.065 |
respect | -0.064 |
rebuild | -0.063 |
Token | a |
Token ID | 257 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
tted | +0.041 |
deserve | +0.039 |
construct | +0.038 |
respect | +0.037 |
ws | +0.037 |
Neg logit contributions | |
ittens | -0.032 |
ls | -0.030 |
ogging | -0.026 |
ule | -0.026 |
ip | -0.025 |
Token | big |
Token ID | 1263 |
Feature activation | -5.04e-01 |
Pos logit contributions | |
tted | +0.005 |
deserve | +0.005 |
construct | +0.005 |
respect | +0.005 |
make | +0.005 |
Neg logit contributions | |
ittens | -0.004 |
ls | -0.004 |
ogging | -0.004 |
ries | -0.003 |
ule | -0.003 |
Token | p |
Token ID | 279 |
Feature activation | -1.26e+01 |
Pos logit contributions | |
ittens | +0.014 |
ls | +0.014 |
ogging | +0.012 |
ule | +0.012 |
ries | +0.012 |
Neg logit contributions | |
tted | -0.013 |
deserve | -0.013 |
construct | -0.012 |
respect | -0.012 |
make | -0.012 |
Token | endant |
Token ID | 23048 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
ittens | +0.242 |
ls | +0.210 |
she | +0.203 |
ogging | +0.201 |
ule | +0.200 |
Neg logit contributions | |
tted | -0.432 |
deserve | -0.395 |
ws | -0.391 |
construct | -0.383 |
respect | -0.378 |
Token | . |
Token ID | 13 |
Feature activation | -4.73e-01 |
Pos logit contributions | |
ittens | +0.056 |
ls | +0.050 |
ogging | +0.045 |
ule | +0.044 |
shrimp | +0.044 |
Neg logit contributions | |
tted | -0.074 |
deserve | -0.069 |
construct | -0.067 |
respect | -0.065 |
make | -0.065 |
Token | Sara |
Token ID | 24799 |
Feature activation | +7.50e-02 |
Pos logit contributions | |
ittens | +0.014 |
ls | +0.013 |
ule | +0.012 |
ogging | +0.012 |
ries | +0.011 |
Neg logit contributions | |
tted | -0.013 |
deserve | -0.012 |
construct | -0.011 |
respect | -0.011 |
ws | -0.011 |
Token | put |
Token ID | 1234 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
tted | +0.002 |
deserve | +0.002 |
construct | +0.002 |
rebuild | +0.002 |
ws | +0.002 |
Neg logit contributions | |
ittens | -0.002 |
ls | -0.002 |
shrimp | -0.001 |
ip | -0.001 |
ogging | -0.001 |
Token | it |
Token ID | 340 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
tted | +0.009 |
deserve | +0.009 |
construct | +0.008 |
ws | +0.008 |
respect | +0.008 |
Neg logit contributions | |
ittens | -0.005 |
ls | -0.005 |
ogging | -0.004 |
ule | -0.004 |
shrimp | -0.004 |
Token | ground |
Token ID | 2323 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
tted | +0.111 |
deserve | +0.105 |
construct | +0.104 |
rebuild | +0.099 |
ws | +0.098 |
Neg logit contributions | |
ittens | -0.093 |
ls | -0.089 |
ogging | -0.075 |
shrimp | -0.075 |
ip | -0.075 |
Token | and |
Token ID | 290 |
Feature activation | -7.86e+00 |
Pos logit contributions | |
ittens | +0.025 |
ls | +0.023 |
shrimp | +0.016 |
ogging | +0.016 |
ip | +0.016 |
Neg logit contributions | |
tted | -0.073 |
deserve | -0.070 |
construct | -0.069 |
rebuild | -0.067 |
ws | -0.067 |
Token | made |
Token ID | 925 |
Feature activation | +2.73e+00 |
Pos logit contributions | |
ittens | +0.192 |
ls | +0.176 |
ogging | +0.153 |
ule | +0.153 |
ip | +0.149 |
Neg logit contributions | |
tted | -0.248 |
deserve | -0.238 |
construct | -0.235 |
rebuild | -0.227 |
ws | -0.223 |
Token | a |
Token ID | 257 |
Feature activation | -8.17e-02 |
Pos logit contributions | |
tted | +0.088 |
deserve | +0.084 |
construct | +0.081 |
ws | +0.079 |
rebuild | +0.079 |
Neg logit contributions | |
ittens | -0.063 |
ls | -0.060 |
she | -0.053 |
strawberries | -0.051 |
shrimp | -0.050 |
Token | big |
Token ID | 1263 |
Feature activation | -8.47e-01 |
Pos logit contributions | |
ittens | +0.002 |
ls | +0.002 |
shrimp | +0.001 |
ogging | +0.001 |
ip | +0.001 |
Neg logit contributions | |
tted | -0.003 |
deserve | -0.003 |
construct | -0.003 |
ws | -0.003 |
rebuild | -0.002 |
Token | p |
Token ID | 279 |
Feature activation | -1.26e+01 |
Pos logit contributions | |
ittens | +0.021 |
ls | +0.020 |
ule | +0.017 |
ip | +0.017 |
ogging | +0.017 |
Neg logit contributions | |
tted | -0.025 |
deserve | -0.024 |
construct | -0.024 |
rebuild | -0.024 |
respect | -0.023 |
Token | uddle |
Token ID | 24500 |
Feature activation | -4.82e+00 |
Pos logit contributions | |
ittens | +0.205 |
ls | +0.189 |
shrimp | +0.157 |
she | +0.155 |
strawberries | +0.154 |
Neg logit contributions | |
tted | -0.490 |
deserve | -0.458 |
construct | -0.457 |
rebuild | -0.441 |
ws | -0.439 |
Token | . |
Token ID | 13 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
ittens | +0.085 |
ls | +0.084 |
she | +0.071 |
shrimp | +0.066 |
ogging | +0.066 |
Neg logit contributions | |
tted | -0.175 |
deserve | -0.170 |
construct | -0.167 |
ws | -0.163 |
rebuild | -0.161 |
Token |
|
Token ID | 198 |
Feature activation | +3.07e+00 |
Pos logit contributions | |
tted | +0.007 |
deserve | +0.007 |
construct | +0.007 |
rebuild | +0.006 |
respect | +0.006 |
Neg logit contributions | |
ls | -0.009 |
ittens | -0.009 |
shrimp | -0.007 |
ule | -0.007 |
ogging | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
tted | +0.085 |
deserve | +0.079 |
construct | +0.078 |
rebuild | +0.074 |
respect | +0.073 |
Neg logit contributions | |
ittens | -0.085 |
ls | -0.083 |
shrimp | -0.072 |
strawberries | -0.071 |
ogging | -0.071 |
Token | The |
Token ID | 464 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
tted | +0.062 |
deserve | +0.057 |
construct | +0.056 |
rebuild | +0.053 |
respect | +0.052 |
Neg logit contributions | |
ittens | -0.078 |
ls | -0.076 |
shrimp | -0.067 |
ip | -0.066 |
ogging | -0.066 |
Token | , |
Token ID | 11 |
Feature activation | -1.05e+01 |
Pos logit contributions | |
ittens | +0.092 |
ls | +0.088 |
shrimp | +0.072 |
ogging | +0.072 |
ip | +0.072 |
Neg logit contributions | |
tted | -0.139 |
deserve | -0.132 |
construct | -0.131 |
rebuild | -0.126 |
ws | -0.125 |
Token | Tim |
Token ID | 5045 |
Feature activation | +7.20e-01 |
Pos logit contributions | |
ittens | +0.264 |
ls | +0.242 |
ogging | +0.231 |
ip | +0.221 |
ipop | +0.214 |
Neg logit contributions | |
make | -0.303 |
tted | -0.297 |
deserve | -0.295 |
construct | -0.289 |
build | -0.288 |
Token | saw |
Token ID | 2497 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
tted | +0.020 |
deserve | +0.019 |
construct | +0.019 |
rebuild | +0.018 |
ws | +0.017 |
Neg logit contributions | |
ittens | -0.020 |
ls | -0.019 |
ip | -0.017 |
ogging | -0.017 |
shrimp | -0.016 |
Token | a |
Token ID | 257 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
tted | +0.050 |
deserve | +0.047 |
construct | +0.046 |
respect | +0.045 |
rebuild | +0.045 |
Neg logit contributions | |
ittens | -0.067 |
ls | -0.063 |
ip | -0.057 |
ogging | -0.057 |
ipop | -0.056 |
Token | big |
Token ID | 1263 |
Feature activation | -7.61e-01 |
Pos logit contributions | |
ittens | +0.004 |
ls | +0.003 |
ries | +0.003 |
ogging | +0.003 |
ip | +0.003 |
Neg logit contributions | |
tted | -0.003 |
deserve | -0.003 |
construct | -0.003 |
form | -0.003 |
make | -0.003 |
Token | p |
Token ID | 279 |
Feature activation | -1.26e+01 |
Pos logit contributions | |
ittens | +0.022 |
ls | +0.020 |
ip | +0.018 |
uts | +0.018 |
ipop | +0.017 |
Neg logit contributions | |
make | -0.020 |
form | -0.020 |
build | -0.020 |
construct | -0.019 |
will | -0.019 |
Token | uddle |
Token ID | 24500 |
Feature activation | -4.58e+00 |
Pos logit contributions | |
ittens | +0.265 |
Gig | +0.260 |
strawberries | +0.252 |
Bobby | +0.247 |
Nem | +0.245 |
Neg logit contributions | |
tted | -0.393 |
construct | -0.327 |
deserve | -0.325 |
age | -0.324 |
form | -0.321 |
Token | outside |
Token ID | 2354 |
Feature activation | -4.53e-01 |
Pos logit contributions | |
ittens | +0.106 |
ls | +0.099 |
ip | +0.083 |
shrimp | +0.082 |
ule | +0.082 |
Neg logit contributions | |
tted | -0.151 |
deserve | -0.141 |
construct | -0.139 |
rebuild | -0.134 |
ws | -0.132 |
Token | his |
Token ID | 465 |
Feature activation | +5.32e-02 |
Pos logit contributions | |
ittens | +0.010 |
ls | +0.009 |
ip | +0.008 |
shrimp | +0.008 |
ogging | +0.008 |
Neg logit contributions | |
tted | -0.015 |
deserve | -0.014 |
construct | -0.014 |
rebuild | -0.014 |
ws | -0.014 |
Token | house |
Token ID | 2156 |
Feature activation | -7.74e-01 |
Pos logit contributions | |
tted | +0.001 |
construct | +0.001 |
deserve | +0.001 |
rebuild | +0.001 |
respect | +0.001 |
Neg logit contributions | |
ittens | -0.002 |
ls | -0.002 |
ip | -0.001 |
ogging | -0.001 |
ipop | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
ittens | +0.018 |
ls | +0.017 |
ip | +0.015 |
shrimp | +0.014 |
ogging | +0.014 |
Neg logit contributions | |
tted | -0.025 |
deserve | -0.023 |
construct | -0.023 |
rebuild | -0.022 |
respect | -0.022 |
Token | back |
Token ID | 736 |
Feature activation | +5.72e-02 |
Pos logit contributions | |
ittens | +0.081 |
ls | +0.071 |
ule | +0.067 |
ries | +0.066 |
accidents | +0.064 |
Neg logit contributions | |
tted | -0.062 |
construct | -0.053 |
make | -0.051 |
rebuild | -0.051 |
deserve | -0.051 |
Token | to |
Token ID | 284 |
Feature activation | +4.01e+00 |
Pos logit contributions | |
tted | +0.002 |
construct | +0.001 |
deserve | +0.001 |
rebuild | +0.001 |
ws | +0.001 |
Neg logit contributions | |
ittens | -0.002 |
ls | -0.001 |
ule | -0.001 |
ip | -0.001 |
shrimp | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +3.86e+00 |
Pos logit contributions | |
tted | +0.125 |
deserve | +0.120 |
construct | +0.119 |
rebuild | +0.115 |
ws | +0.113 |
Neg logit contributions | |
ittens | -0.098 |
ls | -0.094 |
ip | -0.079 |
ule | -0.077 |
ogging | -0.077 |
Token | park |
Token ID | 3952 |
Feature activation | -2.83e+00 |
Pos logit contributions | |
tted | +0.107 |
deserve | +0.101 |
construct | +0.099 |
rebuild | +0.095 |
respect | +0.094 |
Neg logit contributions | |
ittens | -0.108 |
ls | -0.104 |
ogging | -0.090 |
ip | -0.089 |
ule | -0.088 |
Token | often |
Token ID | 1690 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
ittens | +0.056 |
ls | +0.052 |
ip | +0.040 |
ule | +0.040 |
strawberries | +0.040 |
Neg logit contributions | |
tted | -0.103 |
construct | -0.097 |
deserve | -0.097 |
rebuild | -0.095 |
overcome | -0.093 |
Token | to |
Token ID | 284 |
Feature activation | +4.18e+00 |
Pos logit contributions | |
tted | +0.006 |
deserve | +0.006 |
construct | +0.006 |
rebuild | +0.006 |
ws | +0.006 |
Neg logit contributions | |
ittens | -0.003 |
ls | -0.003 |
shrimp | -0.002 |
ip | -0.002 |
ogging | -0.002 |
Token | cook |
Token ID | 4255 |
Feature activation | +6.38e-01 |
Pos logit contributions | |
tted | +0.106 |
ws | +0.090 |
construct | +0.090 |
deserve | +0.089 |
stations | +0.087 |
Neg logit contributions | |
shrimp | -0.114 |
she | -0.112 |
ittens | -0.112 |
sc | -0.110 |
Gig | -0.107 |
Token | with |
Token ID | 351 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
tted | +0.020 |
deserve | +0.019 |
construct | +0.019 |
rebuild | +0.018 |
respect | +0.018 |
Neg logit contributions | |
ittens | -0.015 |
ls | -0.014 |
ule | -0.012 |
ogging | -0.012 |
ip | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | +3.69e+00 |
Pos logit contributions | |
tted | +0.072 |
deserve | +0.069 |
construct | +0.068 |
respect | +0.066 |
rebuild | +0.066 |
Neg logit contributions | |
ittens | -0.060 |
ls | -0.057 |
ogging | -0.050 |
ule | -0.050 |
ip | -0.048 |
Token | elderly |
Token ID | 13830 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
tted | +0.125 |
deserve | +0.119 |
construct | +0.117 |
rebuild | +0.113 |
respect | +0.112 |
Neg logit contributions | |
ittens | -0.080 |
ls | -0.076 |
ogging | -0.063 |
ip | -0.062 |
ule | -0.062 |
Token | woman |
Token ID | 2415 |
Feature activation | +4.55e-01 |
Pos logit contributions | |
ittens | +0.042 |
ls | +0.042 |
ip | +0.036 |
ogging | +0.035 |
ipop | +0.035 |
Neg logit contributions | |
tted | -0.053 |
deserve | -0.049 |
respect | -0.047 |
construct | -0.046 |
make | -0.046 |
Token | bit |
Token ID | 1643 |
Feature activation | -8.13e+00 |
Pos logit contributions | |
ittens | +0.013 |
ls | +0.012 |
ries | +0.012 |
ogging | +0.011 |
uts | +0.010 |
Neg logit contributions | |
tted | -0.010 |
deserve | -0.010 |
make | -0.010 |
will | -0.010 |
form | -0.010 |
Token | fearful |
Token ID | 26844 |
Feature activation | -4.25e+00 |
Pos logit contributions | |
ittens | +0.201 |
ls | +0.162 |
ries | +0.153 |
ipop | +0.150 |
ule | +0.149 |
Neg logit contributions | |
tted | -0.266 |
deserve | -0.256 |
construct | -0.247 |
make | -0.242 |
rebuild | -0.238 |
Token | but |
Token ID | 475 |
Feature activation | -3.46e+00 |
Pos logit contributions | |
ittens | +0.109 |
ls | +0.094 |
ip | +0.086 |
ipop | +0.082 |
ries | +0.077 |
Neg logit contributions | |
deserve | -0.125 |
make | -0.124 |
respect | -0.122 |
tted | -0.122 |
construct | -0.118 |
Token | wanted |
Token ID | 2227 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
ittens | +0.076 |
ls | +0.066 |
ule | +0.056 |
ries | +0.054 |
ip | +0.053 |
Neg logit contributions | |
tted | -0.118 |
deserve | -0.116 |
respect | -0.113 |
make | -0.113 |
construct | -0.113 |
Token | to |
Token ID | 284 |
Feature activation | +4.49e+00 |
Pos logit contributions | |
tted | +0.066 |
deserve | +0.063 |
construct | +0.063 |
respect | +0.061 |
overcome | +0.060 |
Neg logit contributions | |
ittens | -0.056 |
ls | -0.053 |
ip | -0.045 |
ule | -0.044 |
ogging | -0.044 |
Token | take |
Token ID | 1011 |
Feature activation | +4.08e+00 |
Pos logit contributions | |
tted | +0.098 |
ws | +0.086 |
deserve | +0.080 |
construct | +0.078 |
gen | +0.075 |
Neg logit contributions | |
ittens | -0.145 |
she | -0.141 |
ls | -0.138 |
sc | -0.135 |
ogging | -0.133 |
Token | a |
Token ID | 257 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
tted | +0.100 |
deserve | +0.095 |
construct | +0.091 |
respect | +0.090 |
make | +0.088 |
Neg logit contributions | |
ittens | -0.127 |
ls | -0.121 |
ule | -0.110 |
ipop | -0.106 |
ogging | -0.104 |
Token | look |
Token ID | 804 |
Feature activation | -2.88e+00 |
Pos logit contributions | |
tted | +0.029 |
deserve | +0.028 |
construct | +0.026 |
make | +0.026 |
respect | +0.026 |
Neg logit contributions | |
ittens | -0.041 |
ls | -0.038 |
ries | -0.036 |
ule | -0.034 |
strawberries | -0.033 |
Token | anyway |
Token ID | 6949 |
Feature activation | -4.74e+00 |
Pos logit contributions | |
ittens | +0.076 |
ls | +0.065 |
ule | +0.061 |
shrimp | +0.058 |
ip | +0.057 |
Neg logit contributions | |
tted | -0.083 |
deserve | -0.079 |
respect | -0.077 |
construct | -0.076 |
make | -0.075 |
Token | . |
Token ID | 13 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
ittens | +0.102 |
ls | +0.096 |
shrimp | +0.080 |
ip | +0.079 |
ule | +0.078 |
Neg logit contributions | |
tted | -0.159 |
deserve | -0.152 |
construct | -0.150 |
rebuild | -0.144 |
respect | -0.143 |
Token | Swe |
Token ID | 19372 |
Feature activation | -5.22e+00 |
Pos logit contributions | |
tted | +0.046 |
deserve | +0.043 |
construct | +0.043 |
rebuild | +0.041 |
ws | +0.041 |
Neg logit contributions | |
ittens | -0.040 |
ls | -0.039 |
shrimp | -0.033 |
ogging | -0.033 |
ip | -0.033 |
Token | was |
Token ID | 373 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
tted | +0.023 |
deserve | +0.022 |
construct | +0.022 |
rebuild | +0.021 |
respect | +0.020 |
Neg logit contributions | |
ittens | -0.022 |
ls | -0.021 |
ogging | -0.018 |
ule | -0.018 |
shrimp | -0.018 |
Token | very |
Token ID | 845 |
Feature activation | -8.46e+00 |
Pos logit contributions | |
ittens | +0.033 |
ls | +0.031 |
ip | +0.024 |
ogging | +0.024 |
shrimp | +0.024 |
Neg logit contributions | |
tted | -0.069 |
deserve | -0.066 |
construct | -0.065 |
rebuild | -0.063 |
ws | -0.062 |
Token | excited |
Token ID | 6568 |
Feature activation | -5.06e+00 |
Pos logit contributions | |
ittens | +0.175 |
ls | +0.158 |
ries | +0.134 |
ip | +0.129 |
ule | +0.127 |
Neg logit contributions | |
tted | -0.308 |
deserve | -0.293 |
construct | -0.280 |
respect | -0.278 |
overcome | -0.277 |
Token | and |
Token ID | 290 |
Feature activation | -8.19e+00 |
Pos logit contributions | |
ittens | +0.122 |
ls | +0.115 |
ip | +0.098 |
ule | +0.091 |
ipop | +0.090 |
Neg logit contributions | |
make | -0.152 |
deserve | -0.151 |
respect | -0.150 |
construct | -0.149 |
tted | -0.146 |
Token | wore |
Token ID | 12408 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
ittens | +0.203 |
ls | +0.187 |
ule | +0.156 |
ogging | +0.154 |
ries | +0.153 |
Neg logit contributions | |
tted | -0.262 |
deserve | -0.255 |
construct | -0.247 |
make | -0.243 |
overcome | -0.240 |
Token | the |
Token ID | 262 |
Feature activation | +4.21e+00 |
Pos logit contributions | |
tted | +0.064 |
deserve | +0.061 |
construct | +0.061 |
rebuild | +0.059 |
ws | +0.059 |
Neg logit contributions | |
ittens | -0.030 |
ls | -0.028 |
she | -0.023 |
strawberries | -0.023 |
shrimp | -0.022 |
Token | whistle |
Token ID | 16121 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
tted | +0.137 |
deserve | +0.126 |
construct | +0.126 |
rebuild | +0.124 |
ws | +0.124 |
Neg logit contributions | |
ittens | -0.097 |
ls | -0.089 |
shrimp | -0.086 |
hungry | -0.080 |
ule | -0.080 |
Token | around |
Token ID | 1088 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
ittens | +0.030 |
ls | +0.029 |
ip | +0.023 |
shrimp | +0.023 |
ogging | +0.023 |
Neg logit contributions | |
tted | -0.050 |
deserve | -0.047 |
construct | -0.047 |
rebuild | -0.045 |
ws | -0.045 |
Token | her |
Token ID | 607 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
tted | +0.076 |
construct | +0.074 |
deserve | +0.072 |
rebuild | +0.071 |
respect | +0.070 |
Neg logit contributions | |
ittens | -0.056 |
ls | -0.053 |
ule | -0.044 |
ip | -0.044 |
ogging | -0.043 |
Token | neck |
Token ID | 7393 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
tted | +0.052 |
construct | +0.049 |
deserve | +0.049 |
rebuild | +0.047 |
respect | +0.046 |
Neg logit contributions | |
ittens | -0.043 |
ls | -0.042 |
ogging | -0.035 |
shrimp | -0.035 |
ip | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
ittens | +0.023 |
ls | +0.022 |
shrimp | +0.017 |
ogging | +0.017 |
ip | +0.017 |
Neg logit contributions | |
tted | -0.038 |
deserve | -0.036 |
construct | -0.036 |
rebuild | -0.035 |
ws | -0.034 |
Token | the |
Token ID | 262 |
Feature activation | +3.65e+00 |
Pos logit contributions | |
ittens | +0.007 |
ls | +0.007 |
ule | +0.006 |
ogging | +0.006 |
shrimp | +0.006 |
Neg logit contributions | |
tted | -0.011 |
deserve | -0.010 |
respect | -0.010 |
ws | -0.010 |
make | -0.010 |
Token | cage |
Token ID | 17186 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
tted | +0.125 |
deserve | +0.118 |
construct | +0.117 |
ws | +0.113 |
rebuild | +0.112 |
Neg logit contributions | |
ittens | -0.078 |
ls | -0.075 |
sc | -0.063 |
shrimp | -0.062 |
strawberries | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | +8.40e-02 |
Pos logit contributions | |
ittens | +0.026 |
ls | +0.026 |
she | +0.020 |
ogging | +0.019 |
ip | +0.018 |
Neg logit contributions | |
tted | -0.067 |
deserve | -0.065 |
construct | -0.064 |
ws | -0.062 |
rebuild | -0.062 |
Token | She |
Token ID | 1375 |
Feature activation | +2.94e+00 |
Pos logit contributions | |
tted | +0.003 |
construct | +0.003 |
deserve | +0.003 |
rebuild | +0.002 |
ws | +0.002 |
Neg logit contributions | |
ittens | -0.002 |
ls | -0.002 |
she | -0.002 |
ogging | -0.002 |
ip | -0.002 |
Token | wanted |
Token ID | 2227 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
tted | +0.087 |
deserve | +0.083 |
construct | +0.081 |
rebuild | +0.078 |
ws | +0.078 |
Neg logit contributions | |
ittens | -0.076 |
ls | -0.072 |
ogging | -0.064 |
ule | -0.062 |
shrimp | -0.062 |
Token | to |
Token ID | 284 |
Feature activation | +4.77e+00 |
Pos logit contributions | |
tted | +0.074 |
deserve | +0.072 |
construct | +0.071 |
respect | +0.070 |
rebuild | +0.069 |
Neg logit contributions | |
ittens | -0.058 |
ls | -0.054 |
ule | -0.046 |
ogging | -0.046 |
ip | -0.046 |
Token | go |
Token ID | 467 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
tted | +0.119 |
construct | +0.099 |
ws | +0.098 |
gen | +0.096 |
deserve | +0.095 |
Neg logit contributions | |
ittens | -0.139 |
sc | -0.136 |
Twe | -0.135 |
she | -0.133 |
ls | -0.133 |
Token | back |
Token ID | 736 |
Feature activation | +6.36e-01 |
Pos logit contributions | |
ittens | +0.039 |
ls | +0.038 |
shrimp | +0.033 |
ip | +0.033 |
ogging | +0.033 |
Neg logit contributions | |
tted | -0.039 |
deserve | -0.037 |
construct | -0.036 |
rebuild | -0.035 |
ws | -0.034 |
Token | to |
Token ID | 284 |
Feature activation | +4.56e+00 |
Pos logit contributions | |
tted | +0.020 |
deserve | +0.019 |
construct | +0.019 |
rebuild | +0.018 |
respect | +0.018 |
Neg logit contributions | |
ittens | -0.016 |
ls | -0.014 |
ip | -0.013 |
ule | -0.013 |
shrimp | -0.012 |
Token | her |
Token ID | 607 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
tted | +0.157 |
deserve | +0.148 |
construct | +0.146 |
rebuild | +0.140 |
ws | +0.140 |
Neg logit contributions | |
ittens | -0.097 |
ls | -0.093 |
shrimp | -0.077 |
she | -0.076 |
ogging | -0.076 |
Token | place |
Token ID | 1295 |
Feature activation | -5.89e-01 |
Pos logit contributions | |
tted | +0.069 |
construct | +0.067 |
rebuild | +0.067 |
deserve | +0.065 |
respect | +0.064 |
Neg logit contributions | |
ittens | -0.084 |
ls | -0.078 |
ogging | -0.073 |
ip | -0.071 |
ule | -0.070 |
Token | a |
Token ID | 257 |
Feature activation | +8.02e-01 |
Pos logit contributions | |
ittens | +0.211 |
ls | +0.205 |
she | +0.182 |
shrimp | +0.177 |
strawberries | +0.175 |
Neg logit contributions | |
tted | -0.242 |
deserve | -0.224 |
construct | -0.221 |
ws | -0.209 |
rebuild | -0.209 |
Token | marker |
Token ID | 18364 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
tted | +0.023 |
deserve | +0.022 |
construct | +0.021 |
ws | +0.020 |
rebuild | +0.020 |
Neg logit contributions | |
ittens | -0.022 |
ls | -0.020 |
shrimp | -0.018 |
ip | -0.018 |
ogging | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
ittens | +0.028 |
ls | +0.026 |
she | +0.021 |
ip | +0.021 |
ipop | +0.021 |
Neg logit contributions | |
tted | -0.059 |
deserve | -0.057 |
construct | -0.056 |
rebuild | -0.054 |
ws | -0.053 |
Token | He |
Token ID | 679 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
tted | +0.006 |
deserve | +0.006 |
construct | +0.005 |
rebuild | +0.005 |
respect | +0.005 |
Neg logit contributions | |
ittens | -0.007 |
ls | -0.007 |
strawberries | -0.006 |
ries | -0.006 |
ule | -0.006 |
Token | taped |
Token ID | 36114 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
tted | +0.048 |
deserve | +0.046 |
construct | +0.045 |
rebuild | +0.044 |
ws | +0.043 |
Neg logit contributions | |
ittens | -0.032 |
ls | -0.030 |
ogging | -0.026 |
ule | -0.026 |
strawberries | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | +3.88e+00 |
Pos logit contributions | |
tted | +0.085 |
deserve | +0.081 |
construct | +0.080 |
rebuild | +0.077 |
ws | +0.075 |
Neg logit contributions | |
ittens | -0.051 |
ls | -0.050 |
ip | -0.041 |
shrimp | -0.040 |
she | -0.040 |
Token | wheel |
Token ID | 7825 |
Feature activation | -2.91e+00 |
Pos logit contributions | |
tted | +0.121 |
deserve | +0.116 |
construct | +0.114 |
rebuild | +0.110 |
overcome | +0.108 |
Neg logit contributions | |
ittens | -0.089 |
ls | -0.084 |
sc | -0.076 |
peanuts | -0.072 |
strawberries | -0.072 |
Token | back |
Token ID | 736 |
Feature activation | -3.54e-01 |
Pos logit contributions | |
ittens | +0.050 |
ls | +0.049 |
she | +0.040 |
ip | +0.040 |
ipop | +0.038 |
Neg logit contributions | |
tted | -0.107 |
deserve | -0.105 |
construct | -0.104 |
ws | -0.099 |
rebuild | -0.099 |
Token | on |
Token ID | 319 |
Feature activation | +3.78e+00 |
Pos logit contributions | |
ittens | +0.008 |
ls | +0.007 |
ule | +0.007 |
shrimp | +0.007 |
ogging | +0.007 |
Neg logit contributions | |
tted | -0.011 |
construct | -0.011 |
deserve | -0.010 |
rebuild | -0.010 |
ws | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | -7.74e+00 |
Pos logit contributions | |
tted | +0.113 |
deserve | +0.107 |
construct | +0.106 |
rebuild | +0.102 |
ws | +0.100 |
Neg logit contributions | |
ittens | -0.097 |
ls | -0.093 |
ogging | -0.080 |
ule | -0.080 |
ip | -0.078 |
Token | colored |
Token ID | 16396 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
ittens | +0.198 |
ls | +0.194 |
she | +0.165 |
shrimp | +0.164 |
ip | +0.164 |
Neg logit contributions | |
tted | -0.230 |
deserve | -0.215 |
construct | -0.211 |
rebuild | -0.201 |
ws | -0.200 |
Token | my |
Token ID | 616 |
Feature activation | +3.47e+00 |
Pos logit contributions | |
tted | +0.106 |
construct | +0.090 |
ws | +0.090 |
deserve | +0.089 |
gen | +0.083 |
Neg logit contributions | |
ittens | -0.168 |
ls | -0.161 |
shrimp | -0.152 |
she | -0.152 |
Gig | -0.151 |
Token | grandma |
Token ID | 49890 |
Feature activation | +3.64e-01 |
Pos logit contributions | |
tted | +0.077 |
deserve | +0.076 |
construct | +0.075 |
rebuild | +0.075 |
will | +0.073 |
Neg logit contributions | |
ittens | -0.111 |
ls | -0.105 |
ip | -0.094 |
ipop | -0.093 |
ogging | -0.093 |
Token | 's |
Token ID | 338 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
tted | +0.011 |
deserve | +0.010 |
construct | +0.010 |
rebuild | +0.010 |
ws | +0.010 |
Neg logit contributions | |
ittens | -0.009 |
ls | -0.009 |
shrimp | -0.007 |
ogging | -0.007 |
ip | -0.007 |
Token | house |
Token ID | 2156 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
tted | +0.035 |
construct | +0.033 |
deserve | +0.032 |
rebuild | +0.032 |
will | +0.032 |
Neg logit contributions | |
ittens | -0.057 |
ls | -0.056 |
ip | -0.050 |
ipop | -0.049 |
ogging | -0.049 |
Token | to |
Token ID | 284 |
Feature activation | +4.36e+00 |
Pos logit contributions | |
ittens | +0.022 |
ls | +0.021 |
shrimp | +0.017 |
ogging | +0.017 |
ip | +0.016 |
Neg logit contributions | |
tted | -0.038 |
deserve | -0.036 |
construct | -0.036 |
rebuild | -0.034 |
ws | -0.034 |
Token | give |
Token ID | 1577 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
tted | +0.102 |
ws | +0.091 |
construct | +0.085 |
gen | +0.082 |
deserve | +0.082 |
Neg logit contributions | |
ittens | -0.131 |
she | -0.130 |
Twe | -0.128 |
Gig | -0.126 |
sc | -0.124 |
Token | her |
Token ID | 607 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
tted | +0.060 |
deserve | +0.058 |
construct | +0.056 |
rebuild | +0.054 |
ws | +0.054 |
Neg logit contributions | |
ittens | -0.040 |
ls | -0.038 |
ogging | -0.031 |
ule | -0.031 |
ip | -0.031 |
Token | some |
Token ID | 617 |
Feature activation | +5.08e+00 |
Pos logit contributions | |
tted | +0.073 |
deserve | +0.069 |
construct | +0.068 |
rebuild | +0.065 |
ws | +0.064 |
Neg logit contributions | |
ittens | -0.065 |
ls | -0.062 |
shrimp | -0.053 |
ogging | -0.053 |
ip | -0.053 |
Token | cookies |
Token ID | 14746 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
tted | +0.150 |
deserve | +0.147 |
respect | +0.141 |
construct | +0.138 |
make | +0.137 |
Neg logit contributions | |
ittens | -0.123 |
ls | -0.120 |
ule | -0.106 |
ogging | -0.104 |
ip | -0.101 |
Token | ." |
Token ID | 526 |
Feature activation | +4.70e-01 |
Pos logit contributions | |
tted | +0.035 |
deserve | +0.034 |
construct | +0.032 |
respect | +0.031 |
rebuild | +0.031 |
Neg logit contributions | |
ittens | -0.030 |
ls | -0.029 |
ogging | -0.025 |
ule | -0.025 |
ip | -0.024 |
Token | |
Token ID | 220 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
tted | +0.011 |
deserve | +0.011 |
construct | +0.010 |
rebuild | +0.010 |
make | +0.010 |
Neg logit contributions | |
ls | -0.015 |
ittens | -0.014 |
shrimp | -0.013 |
strawberries | -0.013 |
ule | -0.013 |
Token | Mom |
Token ID | 11254 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
tted | +0.007 |
deserve | +0.006 |
construct | +0.006 |
rebuild | +0.006 |
make | +0.006 |
Neg logit contributions | |
ls | -0.011 |
ittens | -0.010 |
shrimp | -0.009 |
ule | -0.009 |
ries | -0.008 |
Token | smiled |
Token ID | 13541 |
Feature activation | -3.35e+00 |
Pos logit contributions | |
ittens | +0.035 |
ls | +0.033 |
shrimp | +0.027 |
ogging | +0.027 |
peanuts | +0.027 |
Neg logit contributions | |
tted | -0.048 |
deserve | -0.046 |
construct | -0.046 |
rebuild | -0.044 |
make | -0.044 |
Token | and |
Token ID | 290 |
Feature activation | -8.36e+00 |
Pos logit contributions | |
ittens | +0.070 |
ls | +0.064 |
ule | +0.053 |
ries | +0.052 |
ip | +0.050 |
Neg logit contributions | |
tted | -0.117 |
deserve | -0.112 |
respect | -0.110 |
make | -0.108 |
rebuild | -0.108 |
Token | said |
Token ID | 531 |
Feature activation | -7.67e-01 |
Pos logit contributions | |
ittens | +0.210 |
ls | +0.181 |
ogging | +0.158 |
ule | +0.157 |
ries | +0.151 |
Neg logit contributions | |
tted | -0.255 |
deserve | -0.252 |
make | -0.249 |
will | -0.235 |
construct | -0.235 |
Token | , |
Token ID | 11 |
Feature activation | -9.35e+00 |
Pos logit contributions | |
ittens | +0.013 |
ls | +0.012 |
ule | +0.009 |
ogging | +0.009 |
ries | +0.009 |
Neg logit contributions | |
tted | -0.030 |
deserve | -0.028 |
respect | -0.028 |
make | -0.027 |
construct | -0.027 |
Token | " |
Token ID | 366 |
Feature activation | +3.20e+00 |
Pos logit contributions | |
ls | +0.259 |
ittens | +0.246 |
ries | +0.212 |
ogging | +0.209 |
shrimp | +0.206 |
Neg logit contributions | |
tted | -0.260 |
deserve | -0.249 |
make | -0.243 |
construct | -0.238 |
rebuild | -0.230 |
Token | Congratulations |
Token ID | 45048 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
tted | +0.087 |
construct | +0.086 |
rebuild | +0.085 |
ws | +0.084 |
build | +0.077 |
Neg logit contributions | |
ittens | -0.084 |
she | -0.081 |
strawberries | -0.077 |
ls | -0.074 |
strawberry | -0.073 |
Token | , |
Token ID | 11 |
Feature activation | -8.51e+00 |
Pos logit contributions | |
ittens | +0.040 |
ls | +0.035 |
ule | +0.029 |
ogging | +0.028 |
shrimp | +0.027 |
Neg logit contributions | |
tted | -0.079 |
deserve | -0.078 |
construct | -0.073 |
respect | -0.073 |
will | -0.073 |
Token | honey |
Token ID | 12498 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
ittens | +0.192 |
ls | +0.184 |
ule | +0.151 |
ogging | +0.150 |
ip | +0.149 |
Neg logit contributions | |
tted | -0.277 |
deserve | -0.271 |
construct | -0.261 |
respect | -0.254 |
rebuild | -0.254 |
Token | . |
Token ID | 13 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
ittens | +0.044 |
ls | +0.043 |
she | +0.036 |
ule | +0.035 |
ipop | +0.034 |
Neg logit contributions | |
tted | -0.095 |
construct | -0.090 |
deserve | -0.090 |
ws | -0.088 |
rebuild | -0.087 |
Token | I |
Token ID | 314 |
Feature activation | +4.94e+00 |
Pos logit contributions | |
tted | +0.003 |
construct | +0.003 |
rebuild | +0.003 |
ws | +0.003 |
overcome | +0.003 |
Neg logit contributions | |
ittens | -0.003 |
ls | -0.002 |
ip | -0.002 |
Gig | -0.002 |
she | -0.002 |
Token | to |
Token ID | 284 |
Feature activation | +4.52e+00 |
Pos logit contributions | |
ittens | +0.025 |
ls | +0.023 |
ule | +0.022 |
ries | +0.021 |
strawberries | +0.021 |
Neg logit contributions | |
tted | -0.018 |
deserve | -0.016 |
respect | -0.015 |
construct | -0.014 |
rebuild | -0.014 |
Token | get |
Token ID | 651 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
tted | +0.128 |
deserve | +0.124 |
construct | +0.124 |
rebuild | +0.121 |
respect | +0.119 |
Neg logit contributions | |
ittens | -0.122 |
ls | -0.117 |
ip | -0.100 |
ule | -0.099 |
ipop | -0.098 |
Token | a |
Token ID | 257 |
Feature activation | +7.05e-01 |
Pos logit contributions | |
tted | +0.003 |
deserve | +0.003 |
construct | +0.003 |
respect | +0.003 |
rebuild | +0.003 |
Neg logit contributions | |
ittens | -0.003 |
ls | -0.003 |
ule | -0.003 |
ogging | -0.003 |
ip | -0.003 |
Token | spoon |
Token ID | 24556 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
tted | +0.019 |
deserve | +0.018 |
construct | +0.017 |
rebuild | +0.017 |
respect | +0.017 |
Neg logit contributions | |
ittens | -0.020 |
ls | -0.019 |
ries | -0.017 |
ule | -0.017 |
ip | -0.016 |
Token | from |
Token ID | 422 |
Feature activation | -6.93e-01 |
Pos logit contributions | |
ittens | +0.052 |
ls | +0.050 |
shrimp | +0.040 |
strawberries | +0.040 |
ule | +0.040 |
Neg logit contributions | |
tted | -0.077 |
deserve | -0.072 |
construct | -0.071 |
rebuild | -0.070 |
respect | -0.068 |
Token | the |
Token ID | 262 |
Feature activation | +3.60e+00 |
Pos logit contributions | |
ittens | +0.020 |
ls | +0.018 |
ule | +0.017 |
ogging | +0.017 |
ip | +0.017 |
Neg logit contributions | |
respect | -0.017 |
deserve | -0.017 |
make | -0.017 |
rebuild | -0.017 |
construct | -0.017 |
Token | drawer |
Token ID | 33451 |
Feature activation | -2.91e-01 |
Pos logit contributions | |
tted | +0.111 |
deserve | +0.105 |
construct | +0.103 |
ws | +0.100 |
rebuild | +0.099 |
Neg logit contributions | |
ittens | -0.087 |
ls | -0.084 |
shrimp | -0.074 |
strawberries | -0.072 |
ip | -0.071 |
Token | . |
Token ID | 13 |
Feature activation | +3.30e-01 |
Pos logit contributions | |
ittens | +0.005 |
ls | +0.005 |
ogging | +0.004 |
shrimp | +0.004 |
ip | +0.004 |
Neg logit contributions | |
tted | -0.011 |
deserve | -0.010 |
construct | -0.010 |
rebuild | -0.010 |
ws | -0.010 |
Token | She |
Token ID | 1375 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
tted | +0.007 |
deserve | +0.007 |
respect | +0.006 |
Use | +0.006 |
construct | +0.006 |
Neg logit contributions | |
ls | -0.011 |
ittens | -0.011 |
ries | -0.009 |
shrimp | -0.009 |
ogging | -0.009 |
Token | was |
Token ID | 373 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
tted | +0.063 |
deserve | +0.060 |
construct | +0.058 |
make | +0.057 |
will | +0.057 |
Neg logit contributions | |
ittens | -0.062 |
ls | -0.056 |
ogging | -0.053 |
ule | -0.051 |
ries | -0.049 |
Token | ready |
Token ID | 3492 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
ittens | +0.038 |
ls | +0.037 |
ogging | +0.029 |
ule | +0.029 |
ip | +0.029 |
Neg logit contributions | |
tted | -0.065 |
deserve | -0.062 |
construct | -0.061 |
rebuild | -0.060 |
overcome | -0.059 |
Token | a |
Token ID | 257 |
Feature activation | -6.13e-01 |
Pos logit contributions | |
ittens | +0.049 |
ls | +0.048 |
shrimp | +0.042 |
she | +0.042 |
strawberries | +0.042 |
Neg logit contributions | |
tted | -0.054 |
deserve | -0.051 |
construct | -0.050 |
ws | -0.048 |
rebuild | -0.048 |
Token | fish |
Token ID | 5916 |
Feature activation | -9.94e-01 |
Pos logit contributions | |
ittens | +0.017 |
ls | +0.016 |
ries | +0.014 |
ogging | +0.014 |
ip | +0.014 |
Neg logit contributions | |
tted | -0.017 |
deserve | -0.016 |
construct | -0.016 |
form | -0.016 |
make | -0.016 |
Token | jumping |
Token ID | 14284 |
Feature activation | -2.43e+00 |
Pos logit contributions | |
ittens | +0.020 |
ls | +0.019 |
ip | +0.015 |
strawberries | +0.015 |
ogging | +0.015 |
Neg logit contributions | |
tted | -0.035 |
deserve | -0.033 |
construct | -0.033 |
rebuild | -0.032 |
respect | -0.031 |
Token | out |
Token ID | 503 |
Feature activation | +5.36e-01 |
Pos logit contributions | |
ittens | +0.066 |
ls | +0.058 |
ule | +0.052 |
ip | +0.052 |
ogging | +0.051 |
Neg logit contributions | |
tted | -0.073 |
deserve | -0.067 |
construct | -0.066 |
ws | -0.063 |
rebuild | -0.063 |
Token | of |
Token ID | 286 |
Feature activation | +4.14e+00 |
Pos logit contributions | |
tted | +0.016 |
deserve | +0.016 |
construct | +0.015 |
rebuild | +0.015 |
respect | +0.015 |
Neg logit contributions | |
ittens | -0.014 |
ls | -0.012 |
ip | -0.011 |
ule | -0.011 |
ipop | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | +3.52e+00 |
Pos logit contributions | |
tted | +0.126 |
deserve | +0.120 |
construct | +0.118 |
rebuild | +0.113 |
ws | +0.112 |
Neg logit contributions | |
ittens | -0.104 |
ls | -0.099 |
ip | -0.084 |
ogging | -0.084 |
ule | -0.084 |
Token | water |
Token ID | 1660 |
Feature activation | -3.39e+00 |
Pos logit contributions | |
tted | +0.107 |
deserve | +0.100 |
construct | +0.099 |
rebuild | +0.096 |
ws | +0.095 |
Neg logit contributions | |
ls | -0.086 |
ittens | -0.086 |
shrimp | -0.076 |
strawberries | -0.073 |
ogging | -0.071 |
Token | ! |
Token ID | 0 |
Feature activation | +7.03e-01 |
Pos logit contributions | |
ittens | +0.069 |
ls | +0.063 |
ip | +0.051 |
ogging | +0.050 |
ule | +0.050 |
Neg logit contributions | |
tted | -0.120 |
deserve | -0.114 |
construct | -0.113 |
rebuild | -0.110 |
ws | -0.108 |
Token | The |
Token ID | 383 |
Feature activation | +3.38e+00 |
Pos logit contributions | |
tted | +0.019 |
deserve | +0.018 |
construct | +0.017 |
rebuild | +0.016 |
overcome | +0.016 |
Neg logit contributions | |
ittens | -0.020 |
ls | -0.019 |
ries | -0.016 |
ogging | -0.016 |
strawberries | -0.016 |
Token | spider |
Token ID | 19230 |
Feature activation | -5.96e-01 |
Pos logit contributions | |
tted | +0.122 |
deserve | +0.116 |
construct | +0.115 |
rebuild | +0.112 |
ws | +0.111 |
Neg logit contributions | |
ittens | -0.065 |
ls | -0.062 |
shrimp | -0.050 |
strawberries | -0.049 |
ogging | -0.049 |
Token | wanted |
Token ID | 2227 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
ittens | +0.018 |
ls | +0.016 |
ogging | +0.015 |
ries | +0.015 |
Gig | +0.015 |
Neg logit contributions | |
tted | -0.015 |
deserve | -0.014 |
construct | -0.014 |
make | -0.014 |
overcome | -0.013 |
Token | Daisy |
Token ID | 40355 |
Feature activation | +3.86e-01 |
Pos logit contributions | |
ittens | +0.093 |
ip | +0.078 |
ogging | +0.078 |
ls | +0.078 |
ule | +0.077 |
Neg logit contributions | |
make | -0.114 |
deserve | -0.109 |
tted | -0.106 |
overcome | -0.106 |
begin | -0.106 |
Token | was |
Token ID | 373 |
Feature activation | -8.20e-01 |
Pos logit contributions | |
tted | +0.009 |
deserve | +0.008 |
construct | +0.008 |
make | +0.008 |
will | +0.008 |
Neg logit contributions | |
ittens | -0.013 |
ls | -0.012 |
ries | -0.011 |
ule | -0.011 |
Gig | -0.011 |
Token | playing |
Token ID | 2712 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
ittens | +0.016 |
ls | +0.016 |
Gig | +0.013 |
ule | +0.013 |
ipop | +0.013 |
Neg logit contributions | |
tted | -0.029 |
overcome | -0.028 |
construct | -0.027 |
build | -0.027 |
rebuild | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | -9.42e+00 |
Pos logit contributions | |
tted | +0.043 |
deserve | +0.040 |
construct | +0.040 |
rebuild | +0.039 |
ws | +0.038 |
Neg logit contributions | |
ittens | -0.022 |
ls | -0.020 |
ule | -0.016 |
ip | -0.016 |
ogging | -0.016 |
Token | she |
Token ID | 673 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
ittens | +0.170 |
ls | +0.146 |
ip | +0.137 |
ogging | +0.133 |
ule | +0.126 |
Neg logit contributions | |
make | -0.337 |
tted | -0.327 |
deserve | -0.327 |
overcome | -0.325 |
construct | -0.318 |
Token | hurt |
Token ID | 5938 |
Feature activation | +2.37e-01 |
Pos logit contributions | |
tted | +0.022 |
deserve | +0.021 |
overcome | +0.021 |
construct | +0.021 |
make | +0.020 |
Neg logit contributions | |
ittens | -0.034 |
ls | -0.030 |
Gig | -0.028 |
ries | -0.028 |
ogging | -0.028 |
Token | herself |
Token ID | 5223 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
tted | +0.006 |
make | +0.006 |
deserve | +0.006 |
respect | +0.006 |
overcome | +0.005 |
Neg logit contributions | |
ls | -0.007 |
ittens | -0.006 |
ip | -0.006 |
ule | -0.005 |
ogging | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +4.38e-02 |
Pos logit contributions | |
tted | +0.058 |
deserve | +0.055 |
construct | +0.054 |
rebuild | +0.053 |
overcome | +0.052 |
Neg logit contributions | |
ittens | -0.043 |
ls | -0.041 |
ule | -0.035 |
shrimp | -0.035 |
ip | -0.034 |
Token | Her |
Token ID | 2332 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
tted | +0.001 |
deserve | +0.001 |
construct | +0.001 |
gen | +0.001 |
make | +0.001 |
Neg logit contributions | |
ittens | -0.001 |
ls | -0.001 |
ip | -0.001 |
ogging | -0.001 |
ule | -0.001 |
Token | mum |
Token ID | 25682 |
Feature activation | -5.46e-01 |
Pos logit contributions | |
tted | +0.011 |
deserve | +0.010 |
make | +0.010 |
construct | +0.010 |
respect | +0.010 |
Neg logit contributions | |
ls | -0.008 |
ittens | -0.008 |
Gig | -0.007 |
ries | -0.007 |
ogging | -0.007 |
Token | rushed |
Token ID | 13999 |
Feature activation | +8.64e-01 |
Pos logit contributions | |
ittens | +0.015 |
ls | +0.014 |
strawberries | +0.012 |
shrimp | +0.012 |
ogging | +0.012 |
Neg logit contributions | |
tted | -0.016 |
deserve | -0.015 |
construct | -0.014 |
rebuild | -0.014 |
make | -0.014 |
Token | to |
Token ID | 284 |
Feature activation | +4.25e+00 |
Pos logit contributions | |
tted | +0.116 |
construct | +0.109 |
deserve | +0.109 |
rebuild | +0.105 |
ws | +0.105 |
Neg logit contributions | |
ittens | -0.056 |
ls | -0.054 |
she | -0.047 |
shrimp | -0.045 |
strawberries | -0.045 |
Token | help |
Token ID | 1037 |
Feature activation | +9.90e-01 |
Pos logit contributions | |
tted | +0.114 |
ws | +0.100 |
construct | +0.098 |
deserve | +0.097 |
gen | +0.093 |
Neg logit contributions | |
ittens | -0.116 |
she | -0.115 |
ls | -0.107 |
sc | -0.106 |
he | -0.104 |
Token | . |
Token ID | 13 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
tted | +0.038 |
deserve | +0.035 |
construct | +0.035 |
ws | +0.035 |
gen | +0.033 |
Neg logit contributions | |
ittens | -0.017 |
ls | -0.016 |
she | -0.014 |
ogging | -0.013 |
shrimp | -0.013 |
Token | The |
Token ID | 383 |
Feature activation | +2.82e+00 |
Pos logit contributions | |
tted | +0.006 |
construct | +0.006 |
deserve | +0.006 |
ws | +0.005 |
rebuild | +0.005 |
Neg logit contributions | |
ittens | -0.004 |
ls | -0.004 |
she | -0.004 |
ogging | -0.004 |
Gig | -0.004 |
Token | judge |
Token ID | 5052 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
tted | +0.079 |
deserve | +0.074 |
construct | +0.073 |
rebuild | +0.070 |
ws | +0.070 |
Neg logit contributions | |
ittens | -0.078 |
ls | -0.075 |
shrimp | -0.065 |
ogging | -0.064 |
ip | -0.064 |
Token | smiled |
Token ID | 13541 |
Feature activation | -2.83e+00 |
Pos logit contributions | |
ittens | +0.007 |
ls | +0.007 |
shrimp | +0.006 |
ip | +0.006 |
she | +0.006 |
Neg logit contributions | |
tted | -0.010 |
deserve | -0.009 |
construct | -0.009 |
ws | -0.009 |
rebuild | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | -8.14e+00 |
Pos logit contributions | |
ittens | +0.043 |
ls | +0.039 |
ogging | +0.029 |
ip | +0.029 |
ule | +0.029 |
Neg logit contributions | |
tted | -0.114 |
deserve | -0.110 |
construct | -0.109 |
rebuild | -0.106 |
ws | -0.104 |
Token | said |
Token ID | 531 |
Feature activation | -3.60e-01 |
Pos logit contributions | |
ittens | +0.176 |
ls | +0.160 |
ogging | +0.132 |
ule | +0.131 |
ip | +0.128 |
Neg logit contributions | |
tted | -0.282 |
deserve | -0.272 |
construct | -0.264 |
rebuild | -0.255 |
respect | -0.254 |
Token | it |
Token ID | 340 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
ittens | +0.005 |
ls | +0.005 |
ule | +0.003 |
ogging | +0.003 |
ip | +0.003 |
Neg logit contributions | |
tted | -0.015 |
deserve | -0.014 |
construct | -0.014 |
rebuild | -0.014 |
respect | -0.014 |
Token | was |
Token ID | 373 |
Feature activation | -7.30e-01 |
Pos logit contributions | |
tted | +0.062 |
deserve | +0.059 |
construct | +0.059 |
ws | +0.058 |
rebuild | +0.057 |
Neg logit contributions | |
ls | -0.043 |
ittens | -0.041 |
she | -0.039 |
ip | -0.035 |
he | -0.035 |
Token | his |
Token ID | 465 |
Feature activation | +8.56e-01 |
Pos logit contributions | |
ittens | +0.016 |
ls | +0.015 |
she | +0.014 |
shrimp | +0.014 |
strawberries | +0.013 |
Neg logit contributions | |
tted | -0.024 |
construct | -0.022 |
deserve | -0.022 |
rebuild | -0.021 |
ws | -0.021 |
Token | € |
Token ID | 26391 |
Feature activation | -2.62e+00 |
Pos logit contributions | |
ittens | +0.064 |
ip | +0.058 |
ogging | +0.058 |
ls | +0.056 |
because | +0.056 |
Neg logit contributions | |
construct | -0.048 |
rebuild | -0.045 |
tted | -0.043 |
ws | -0.043 |
stations | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | -6.48e+00 |
Pos logit contributions | |
ittens | +0.066 |
ls | +0.063 |
ip | +0.055 |
shrimp | +0.053 |
ogging | +0.053 |
Neg logit contributions | |
tted | -0.079 |
construct | -0.075 |
deserve | -0.074 |
rebuild | -0.071 |
ws | -0.071 |
Token | began |
Token ID | 2540 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
ittens | +0.148 |
ls | +0.142 |
ogging | +0.118 |
ule | +0.116 |
ip | +0.116 |
Neg logit contributions | |
tted | -0.211 |
deserve | -0.202 |
construct | -0.198 |
rebuild | -0.191 |
ws | -0.189 |
Token | to |
Token ID | 284 |
Feature activation | +3.78e+00 |
Pos logit contributions | |
tted | +0.045 |
deserve | +0.043 |
construct | +0.042 |
rebuild | +0.041 |
ws | +0.041 |
Neg logit contributions | |
ittens | -0.028 |
ls | -0.026 |
shrimp | -0.022 |
ogging | -0.021 |
ip | -0.021 |
Token | cheer |
Token ID | 14042 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
tted | +0.099 |
deserve | +0.084 |
ws | +0.081 |
construct | +0.079 |
reed | +0.077 |
Neg logit contributions | |
ittens | -0.103 |
sc | -0.100 |
shrimp | -0.099 |
ls | -0.098 |
Gig | -0.096 |
Token | and |
Token ID | 290 |
Feature activation | -7.47e+00 |
Pos logit contributions | |
ittens | +0.034 |
ls | +0.031 |
strawberries | +0.027 |
ule | +0.027 |
ip | +0.027 |
Neg logit contributions | |
tted | -0.045 |
deserve | -0.043 |
construct | -0.042 |
respect | -0.042 |
rebuild | -0.041 |
Token | play |
Token ID | 711 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
ittens | +0.225 |
ls | +0.219 |
shrimp | +0.196 |
strawberries | +0.192 |
ogging | +0.192 |
Neg logit contributions | |
tted | -0.187 |
deserve | -0.171 |
construct | -0.168 |
ws | -0.160 |
rebuild | -0.159 |
Token | in |
Token ID | 287 |
Feature activation | +4.55e+00 |
Pos logit contributions | |
tted | +0.038 |
deserve | +0.037 |
construct | +0.036 |
ws | +0.036 |
rebuild | +0.035 |
Neg logit contributions | |
ls | -0.018 |
ittens | -0.015 |
shrimp | -0.015 |
she | -0.013 |
because | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | +4.34e+00 |
Pos logit contributions | |
tted | +0.159 |
deserve | +0.147 |
construct | +0.144 |
ws | +0.140 |
rebuild | +0.137 |
Neg logit contributions | |
ittens | -0.093 |
ls | -0.091 |
shrimp | -0.081 |
strawberries | -0.077 |
she | -0.076 |
Token | ocean |
Token ID | 9151 |
Feature activation | -3.97e+00 |
Pos logit contributions | |
tted | +0.141 |
deserve | +0.134 |
construct | +0.132 |
rebuild | +0.127 |
ws | +0.126 |
Neg logit contributions | |
ittens | -0.101 |
ls | -0.096 |
shrimp | -0.080 |
ogging | -0.080 |
ip | -0.080 |
Token | . |
Token ID | 13 |
Feature activation | +7.35e-01 |
Pos logit contributions | |
ls | +0.071 |
ittens | +0.069 |
shrimp | +0.057 |
she | +0.055 |
strawberries | +0.054 |
Neg logit contributions | |
tted | -0.147 |
deserve | -0.138 |
construct | -0.138 |
ws | -0.132 |
rebuild | -0.132 |
Token | One |
Token ID | 3198 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
tted | +0.066 |
deserve | +0.061 |
construct | +0.059 |
rebuild | +0.054 |
respect | +0.054 |
Neg logit contributions | |
ittens | -0.128 |
ls | -0.125 |
shrimp | -0.112 |
ip | -0.112 |
ogging | -0.111 |
Token | day |
Token ID | 1110 |
Feature activation | -4.19e+00 |
Pos logit contributions | |
tted | +0.050 |
deserve | +0.048 |
construct | +0.044 |
respect | +0.044 |
make | +0.044 |
Neg logit contributions | |
ittens | -0.058 |
ls | -0.057 |
ries | -0.051 |
ip | -0.049 |
strawberries | -0.048 |
Token | , |
Token ID | 11 |
Feature activation | -9.02e+00 |
Pos logit contributions | |
ittens | +0.098 |
ls | +0.088 |
lee | +0.080 |
actus | +0.079 |
ogging | +0.078 |
Neg logit contributions | |
tted | -0.124 |
make | -0.121 |
respect | -0.121 |
deserve | -0.121 |
gen | -0.118 |
Token | John |
Token ID | 1757 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
ittens | +0.224 |
ls | +0.213 |
ogging | +0.193 |
ip | +0.191 |
ule | +0.176 |
Neg logit contributions | |
deserve | -0.270 |
tted | -0.267 |
construct | -0.263 |
make | -0.262 |
respect | -0.256 |
Token | was |
Token ID | 373 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
ittens | +0.001 |
ls | +0.001 |
ogging | +0.001 |
shrimp | +0.001 |
ip | +0.001 |
Neg logit contributions | |
tted | -0.001 |
deserve | -0.001 |
construct | -0.001 |
rebuild | -0.001 |
ws | -0.001 |
Token | so |
Token ID | 523 |
Feature activation | -7.92e+00 |
Pos logit contributions | |
ittens | +0.021 |
ls | +0.020 |
ogging | +0.015 |
ip | +0.015 |
shrimp | +0.015 |
Neg logit contributions | |
tted | -0.044 |
deserve | -0.042 |
construct | -0.042 |
rebuild | -0.041 |
overcome | -0.040 |
Token | lazy |
Token ID | 16931 |
Feature activation | -4.10e+00 |
Pos logit contributions | |
ittens | +0.139 |
ls | +0.126 |
ries | +0.110 |
ipop | +0.105 |
ip | +0.103 |
Neg logit contributions | |
tted | -0.299 |
deserve | -0.280 |
overcome | -0.276 |
construct | -0.272 |
rebuild | -0.272 |
Token | that |
Token ID | 326 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
ittens | +0.110 |
ls | +0.097 |
ipop | +0.090 |
ries | +0.089 |
strawberries | +0.087 |
Neg logit contributions | |
deserve | -0.106 |
make | -0.106 |
tted | -0.104 |
will | -0.101 |
construct | -0.098 |
Token | he |
Token ID | 339 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
deserve | +0.008 |
tted | +0.008 |
make | +0.008 |
construct | +0.008 |
respect | +0.008 |
Neg logit contributions | |
ittens | -0.003 |
ls | -0.003 |
ip | -0.002 |
ogging | -0.002 |
ries | -0.002 |
Token | didn |
Token ID | 1422 |
Feature activation | -3.35e-01 |
Pos logit contributions | |
ittens | +0.009 |
ls | +0.008 |
ogging | +0.008 |
ries | +0.008 |
ule | +0.007 |
Neg logit contributions | |
tted | -0.009 |
deserve | -0.009 |
construct | -0.008 |
overcome | -0.008 |
make | -0.008 |
Token | 't |
Token ID | 470 |
Feature activation | +4.30e-01 |
Pos logit contributions | |
ittens | +0.010 |
ls | +0.010 |
shrimp | +0.009 |
ip | +0.009 |
ogging | +0.009 |
Neg logit contributions | |
tted | -0.008 |
deserve | -0.008 |
construct | -0.008 |
rebuild | -0.007 |
ws | -0.007 |
Token | be |
Token ID | 307 |
Feature activation | +2.57e-01 |
Pos logit contributions | |
tted | +0.111 |
ws | +0.092 |
irteen | +0.086 |
construct | +0.086 |
deserve | +0.083 |
Neg logit contributions | |
sc | -0.146 |
ittens | -0.145 |
she | -0.141 |
Twe | -0.137 |
he | -0.133 |
Token | gentle |
Token ID | 10296 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
tted | +0.008 |
construct | +0.008 |
deserve | +0.008 |
ws | +0.007 |
rebuild | +0.007 |
Neg logit contributions | |
ittens | -0.006 |
ls | -0.005 |
she | -0.005 |
shrimp | -0.005 |
hungry | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | -8.53e+00 |
Pos logit contributions | |
tted | +0.016 |
deserve | +0.015 |
construct | +0.015 |
ws | +0.014 |
rebuild | +0.014 |
Neg logit contributions | |
ittens | -0.006 |
ls | -0.006 |
shrimp | -0.005 |
ogging | -0.005 |
she | -0.005 |
Token | careful |
Token ID | 8161 |
Feature activation | -2.99e+00 |
Pos logit contributions | |
ittens | +0.204 |
she | +0.202 |
ls | +0.196 |
hungry | +0.185 |
he | +0.185 |
Neg logit contributions | |
tted | -0.254 |
construct | -0.228 |
ws | -0.227 |
deserve | -0.222 |
irteen | -0.210 |
Token | , |
Token ID | 11 |
Feature activation | -9.45e+00 |
Pos logit contributions | |
ittens | +0.057 |
ls | +0.054 |
she | +0.047 |
ogging | +0.043 |
ule | +0.042 |
Neg logit contributions | |
tted | -0.108 |
deserve | -0.103 |
construct | -0.103 |
ws | -0.099 |
rebuild | -0.097 |
Token | but |
Token ID | 475 |
Feature activation | -3.18e+00 |
Pos logit contributions | |
ittens | +0.189 |
ls | +0.183 |
ip | +0.143 |
ule | +0.142 |
ogging | +0.142 |
Neg logit contributions | |
tted | -0.331 |
deserve | -0.325 |
construct | -0.315 |
respect | -0.310 |
rebuild | -0.310 |
Token | instead |
Token ID | 2427 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
ittens | +0.057 |
ls | +0.053 |
ip | +0.041 |
ogging | +0.041 |
ule | +0.040 |
Neg logit contributions | |
tted | -0.119 |
deserve | -0.116 |
construct | -0.114 |
respect | -0.111 |
rebuild | -0.111 |
Token | he |
Token ID | 339 |
Feature activation | +9.76e-01 |
Pos logit contributions | |
tted | +0.035 |
deserve | +0.034 |
construct | +0.033 |
rebuild | +0.032 |
ws | +0.032 |
Neg logit contributions | |
ittens | -0.023 |
ls | -0.022 |
shrimp | -0.018 |
ip | -0.018 |
ogging | -0.018 |
Token | decided |
Token ID | 3066 |
Feature activation | -2.31e+00 |
Pos logit contributions | |
tted | +0.027 |
construct | +0.026 |
deserve | +0.025 |
rebuild | +0.024 |
ws | +0.024 |
Neg logit contributions | |
ittens | -0.026 |
ls | -0.025 |
ip | -0.022 |
she | -0.022 |
shrimp | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | +4.86e+00 |
Pos logit contributions | |
ittens | +0.054 |
ls | +0.050 |
ule | +0.042 |
ip | +0.042 |
shrimp | +0.041 |
Neg logit contributions | |
tted | -0.075 |
deserve | -0.072 |
construct | -0.071 |
rebuild | -0.069 |
respect | -0.069 |
Token | do |
Token ID | 466 |
Feature activation | +4.03e+00 |
Pos logit contributions | |
tted | +0.106 |
ws | +0.092 |
construct | +0.087 |
irteen | +0.084 |
deserve | +0.083 |
Neg logit contributions | |
ittens | -0.152 |
sc | -0.145 |
she | -0.142 |
Twe | -0.139 |
because | -0.138 |
Token | he |
Token ID | 339 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
ittens | +0.021 |
ls | +0.020 |
ogging | +0.015 |
shrimp | +0.015 |
ip | +0.015 |
Neg logit contributions | |
tted | -0.049 |
deserve | -0.047 |
construct | -0.047 |
rebuild | -0.045 |
ws | -0.045 |
Token | was |
Token ID | 373 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
tted | +0.030 |
construct | +0.029 |
deserve | +0.028 |
rebuild | +0.028 |
ws | +0.027 |
Neg logit contributions | |
ittens | -0.025 |
ls | -0.025 |
she | -0.021 |
ip | -0.021 |
shrimp | -0.021 |
Token | only |
Token ID | 691 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
ittens | +0.028 |
ls | +0.026 |
ip | +0.022 |
ogging | +0.022 |
ule | +0.022 |
Neg logit contributions | |
tted | -0.035 |
deserve | -0.034 |
construct | -0.033 |
rebuild | -0.032 |
respect | -0.032 |
Token | visiting |
Token ID | 10013 |
Feature activation | +3.31e+00 |
Pos logit contributions | |
ittens | +0.047 |
ls | +0.045 |
ogging | +0.037 |
ip | +0.037 |
ule | +0.037 |
Neg logit contributions | |
tted | -0.066 |
deserve | -0.063 |
construct | -0.062 |
rebuild | -0.060 |
ws | -0.060 |
Token | , |
Token ID | 11 |
Feature activation | -9.44e+00 |
Pos logit contributions | |
tted | +0.107 |
deserve | +0.101 |
construct | +0.100 |
rebuild | +0.097 |
ws | +0.095 |
Neg logit contributions | |
ittens | -0.077 |
ls | -0.073 |
ip | -0.061 |
ogging | -0.060 |
ule | -0.060 |
Token | and |
Token ID | 290 |
Feature activation | -8.18e+00 |
Pos logit contributions | |
ittens | +0.160 |
ls | +0.151 |
ip | +0.114 |
ogging | +0.113 |
shrimp | +0.112 |
Neg logit contributions | |
tted | -0.363 |
deserve | -0.350 |
construct | -0.346 |
rebuild | -0.337 |
respect | -0.332 |
Token | asked |
Token ID | 1965 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
ittens | +0.144 |
ls | +0.131 |
ogging | +0.101 |
ip | +0.101 |
ule | +0.100 |
Neg logit contributions | |
tted | -0.312 |
deserve | -0.302 |
construct | -0.298 |
rebuild | -0.292 |
respect | -0.287 |
Token | if |
Token ID | 611 |
Feature activation | -6.21e-01 |
Pos logit contributions | |
tted | +0.082 |
deserve | +0.078 |
construct | +0.078 |
rebuild | +0.076 |
respect | +0.075 |
Neg logit contributions | |
ittens | -0.067 |
ls | -0.064 |
ogging | -0.053 |
ip | -0.053 |
ule | -0.053 |
Token | he |
Token ID | 339 |
Feature activation | +7.59e-01 |
Pos logit contributions | |
ittens | +0.008 |
ls | +0.006 |
ogging | +0.005 |
ip | +0.004 |
ule | +0.004 |
Neg logit contributions | |
tted | -0.026 |
deserve | -0.026 |
construct | -0.026 |
respect | -0.025 |
rebuild | -0.025 |
Token | could |
Token ID | 714 |
Feature activation | +3.57e+00 |
Pos logit contributions | |
tted | +0.021 |
deserve | +0.019 |
construct | +0.019 |
rebuild | +0.018 |
ws | +0.018 |
Neg logit contributions | |
ittens | -0.021 |
ls | -0.021 |
shrimp | -0.018 |
ogging | -0.018 |
ip | -0.018 |
Token | stay |
Token ID | 2652 |
Feature activation | -9.90e-01 |
Pos logit contributions | |
tted | +0.070 |
construct | +0.063 |
ws | +0.062 |
deserve | +0.061 |
gen | +0.056 |
Neg logit contributions | |
ittens | -0.122 |
ls | -0.118 |
she | -0.113 |
ogging | -0.108 |
sc | -0.107 |
Token | . |
Token ID | 13 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
ittens | +0.039 |
ls | +0.037 |
shrimp | +0.031 |
ule | +0.030 |
ip | +0.030 |
Neg logit contributions | |
tted | -0.054 |
deserve | -0.052 |
construct | -0.051 |
rebuild | -0.049 |
respect | -0.049 |
Token |
|
Token ID | 198 |
Feature activation | +4.04e+00 |
Pos logit contributions | |
tted | +0.035 |
deserve | +0.033 |
construct | +0.032 |
rebuild | +0.031 |
respect | +0.030 |
Neg logit contributions | |
ittens | -0.035 |
ls | -0.035 |
shrimp | -0.030 |
ip | -0.030 |
strawberries | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | +3.36e+00 |
Pos logit contributions | |
tted | +0.119 |
deserve | +0.114 |
construct | +0.112 |
rebuild | +0.108 |
ws | +0.108 |
Neg logit contributions | |
ittens | -0.105 |
ls | -0.099 |
ogging | -0.085 |
ule | -0.085 |
ip | -0.085 |
Token | One |
Token ID | 3198 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
tted | +0.080 |
deserve | +0.075 |
construct | +0.073 |
ws | +0.070 |
rebuild | +0.070 |
Neg logit contributions | |
ittens | -0.107 |
ls | -0.102 |
ule | -0.091 |
ogging | -0.091 |
ip | -0.090 |
Token | day |
Token ID | 1110 |
Feature activation | -4.17e+00 |
Pos logit contributions | |
tted | +0.049 |
deserve | +0.046 |
construct | +0.044 |
make | +0.044 |
respect | +0.043 |
Neg logit contributions | |
ittens | -0.066 |
ls | -0.065 |
ries | -0.059 |
strawberries | -0.055 |
Twe | -0.055 |
Token | , |
Token ID | 11 |
Feature activation | -9.14e+00 |
Pos logit contributions | |
ittens | +0.080 |
ls | +0.074 |
actus | +0.069 |
ogging | +0.065 |
strawberries | +0.064 |
Neg logit contributions | |
tted | -0.144 |
construct | -0.142 |
make | -0.137 |
rebuild | -0.136 |
build | -0.135 |
Token | while |
Token ID | 981 |
Feature activation | -4.68e+00 |
Pos logit contributions | |
ls | +0.212 |
ittens | +0.211 |
ogging | +0.191 |
ip | +0.186 |
shrimp | +0.180 |
Neg logit contributions | |
construct | -0.275 |
tted | -0.270 |
make | -0.268 |
rebuild | -0.264 |
build | -0.260 |
Token | Andy |
Token ID | 12382 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
actus | +0.115 |
ittens | +0.113 |
ls | +0.112 |
ip | +0.104 |
ears | +0.103 |
Neg logit contributions | |
build | -0.128 |
construct | -0.125 |
make | -0.125 |
rebuild | -0.123 |
overcome | -0.121 |
Token | was |
Token ID | 373 |
Feature activation | -9.89e-01 |
Pos logit contributions | |
ittens | +0.006 |
ls | +0.005 |
shrimp | +0.005 |
strawberries | +0.005 |
ule | +0.005 |
Neg logit contributions | |
tted | -0.006 |
deserve | -0.006 |
construct | -0.006 |
rebuild | -0.005 |
overcome | -0.005 |
Token | playing |
Token ID | 2712 |
Feature activation | +9.74e-01 |
Pos logit contributions | |
ittens | +0.021 |
ls | +0.020 |
ule | +0.015 |
ip | +0.015 |
shrimp | +0.015 |
Neg logit contributions | |
tted | -0.034 |
construct | -0.033 |
deserve | -0.032 |
overcome | -0.032 |
rebuild | -0.032 |
Token | tag |
Token ID | 7621 |
Feature activation | -5.64e-01 |
Pos logit contributions | |
tted | +0.037 |
construct | +0.035 |
deserve | +0.035 |
rebuild | +0.034 |
ws | +0.033 |
Neg logit contributions | |
ittens | -0.017 |
ls | -0.016 |
ip | -0.012 |
ule | -0.012 |
shrimp | -0.012 |
Token | safe |
Token ID | 3338 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
ittens | +0.188 |
ls | +0.174 |
ogging | +0.148 |
ip | +0.147 |
ries | +0.146 |
Neg logit contributions | |
tted | -0.255 |
deserve | -0.244 |
construct | -0.236 |
rebuild | -0.230 |
overcome | -0.228 |
Token | . |
Token ID | 13 |
Feature activation | +3.82e-01 |
Pos logit contributions | |
ittens | +0.023 |
ls | +0.021 |
ip | +0.018 |
ogging | +0.017 |
shrimp | +0.017 |
Neg logit contributions | |
tted | -0.037 |
deserve | -0.035 |
construct | -0.035 |
rebuild | -0.034 |
ws | -0.033 |
Token | He |
Token ID | 679 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
tted | +0.010 |
deserve | +0.010 |
construct | +0.009 |
rebuild | +0.009 |
ws | +0.009 |
Neg logit contributions | |
ittens | -0.011 |
ls | -0.011 |
shrimp | -0.009 |
strawberries | -0.009 |
ogging | -0.009 |
Token | became |
Token ID | 2627 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
tted | +0.041 |
deserve | +0.039 |
construct | +0.038 |
overcome | +0.037 |
rebuild | +0.036 |
Neg logit contributions | |
ittens | -0.044 |
ls | -0.041 |
ogging | -0.037 |
ries | -0.036 |
ule | -0.036 |
Token | brave |
Token ID | 14802 |
Feature activation | -3.59e+00 |
Pos logit contributions | |
ittens | +0.033 |
ls | +0.032 |
shrimp | +0.027 |
ip | +0.027 |
strawberries | +0.027 |
Neg logit contributions | |
tted | -0.047 |
deserve | -0.044 |
construct | -0.044 |
rebuild | -0.042 |
ws | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | -9.19e+00 |
Pos logit contributions | |
ittens | +0.050 |
ls | +0.047 |
shrimp | +0.034 |
ogging | +0.034 |
ip | +0.033 |
Neg logit contributions | |
tted | -0.149 |
deserve | -0.143 |
construct | -0.141 |
rebuild | -0.137 |
ws | -0.136 |
Token | ran |
Token ID | 4966 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
ittens | +0.211 |
ls | +0.208 |
she | +0.180 |
shrimp | +0.176 |
ip | +0.172 |
Neg logit contributions | |
tted | -0.290 |
construct | -0.274 |
deserve | -0.272 |
ws | -0.262 |
rebuild | -0.261 |
Token | back |
Token ID | 736 |
Feature activation | -4.18e-02 |
Pos logit contributions | |
ittens | +0.039 |
ls | +0.036 |
ule | +0.033 |
shrimp | +0.032 |
strawberries | +0.032 |
Neg logit contributions | |
tted | -0.037 |
deserve | -0.034 |
construct | -0.033 |
rebuild | -0.032 |
respect | -0.032 |
Token | to |
Token ID | 284 |
Feature activation | +3.74e+00 |
Pos logit contributions | |
ittens | +0.001 |
ls | +0.001 |
shrimp | +0.001 |
ule | +0.001 |
ipop | +0.001 |
Neg logit contributions | |
tted | -0.001 |
construct | -0.001 |
deserve | -0.001 |
rebuild | -0.001 |
overcome | -0.001 |
Token | play |
Token ID | 711 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
tted | +0.128 |
deserve | +0.122 |
construct | +0.120 |
rebuild | +0.116 |
ws | +0.115 |
Neg logit contributions | |
ittens | -0.081 |
ls | -0.077 |
shrimp | -0.063 |
ogging | -0.063 |
ip | -0.063 |
Token | with |
Token ID | 351 |
Feature activation | +2.74e+00 |
Pos logit contributions | |
tted | +0.068 |
deserve | +0.066 |
construct | +0.064 |
ws | +0.063 |
rebuild | +0.061 |
Neg logit contributions | |
ittens | -0.033 |
ls | -0.032 |
she | -0.026 |
shrimp | -0.026 |
ogging | -0.025 |
Token | hard |
Token ID | 1327 |
Feature activation | -3.27e+00 |
Pos logit contributions | |
ittens | +0.019 |
ls | +0.018 |
ule | +0.014 |
ogging | +0.014 |
shrimp | +0.014 |
Neg logit contributions | |
tted | -0.034 |
deserve | -0.032 |
construct | -0.032 |
rebuild | -0.031 |
ws | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | -7.81e+00 |
Pos logit contributions | |
ittens | +0.066 |
ls | +0.063 |
shrimp | +0.051 |
ogging | +0.050 |
ip | +0.050 |
Neg logit contributions | |
tted | -0.116 |
deserve | -0.110 |
construct | -0.109 |
rebuild | -0.105 |
ws | -0.104 |
Token | finds |
Token ID | 7228 |
Feature activation | +2.93e+00 |
Pos logit contributions | |
ls | +0.207 |
ittens | +0.191 |
she | +0.186 |
he | +0.174 |
strawberries | +0.170 |
Neg logit contributions | |
tted | -0.230 |
deserve | -0.207 |
construct | -0.202 |
rebuild | -0.190 |
gen | -0.189 |
Token | the |
Token ID | 262 |
Feature activation | +3.95e+00 |
Pos logit contributions | |
tted | +0.097 |
deserve | +0.093 |
construct | +0.090 |
rebuild | +0.087 |
respect | +0.084 |
Neg logit contributions | |
ls | -0.067 |
ittens | -0.061 |
shrimp | -0.056 |
she | -0.055 |
strawberries | -0.055 |
Token | head |
Token ID | 1182 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
tted | +0.140 |
construct | +0.129 |
ws | +0.126 |
rebuild | +0.126 |
deserve | +0.124 |
Neg logit contributions | |
ls | -0.075 |
shrimp | -0.069 |
ittens | -0.068 |
hungry | -0.067 |
peanuts | -0.065 |
Token | , |
Token ID | 11 |
Feature activation | -9.87e+00 |
Pos logit contributions | |
ls | +0.040 |
ittens | +0.036 |
she | +0.033 |
ip | +0.031 |
he | +0.030 |
Neg logit contributions | |
tted | -0.077 |
deserve | -0.074 |
construct | -0.074 |
rebuild | -0.071 |
ws | -0.071 |
Token | the |
Token ID | 262 |
Feature activation | +3.29e+00 |
Pos logit contributions | |
ittens | +0.212 |
ls | +0.196 |
ule | +0.164 |
ogging | +0.164 |
ip | +0.162 |
Neg logit contributions | |
tted | -0.336 |
construct | -0.328 |
deserve | -0.326 |
rebuild | -0.317 |
ws | -0.313 |
Token | tail |
Token ID | 7894 |
Feature activation | -8.24e-02 |
Pos logit contributions | |
tted | +0.115 |
construct | +0.106 |
deserve | +0.106 |
ws | +0.104 |
rebuild | +0.103 |
Neg logit contributions | |
ittens | -0.066 |
ls | -0.063 |
shrimp | -0.058 |
hungry | -0.056 |
strawberries | -0.055 |
Token | , |
Token ID | 11 |
Feature activation | -1.02e+01 |
Pos logit contributions | |
ls | +0.002 |
ittens | +0.002 |
she | +0.002 |
shrimp | +0.002 |
he | +0.002 |
Neg logit contributions | |
tted | -0.002 |
construct | -0.002 |
deserve | -0.002 |
rebuild | -0.002 |
ws | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | +3.46e+00 |
Pos logit contributions | |
ittens | +0.244 |
ls | +0.228 |
ogging | +0.196 |
ule | +0.195 |
ip | +0.192 |
Neg logit contributions | |
tted | -0.320 |
construct | -0.307 |
deserve | -0.306 |
rebuild | -0.297 |
ws | -0.295 |
Token | legs |
Token ID | 7405 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
tted | +0.118 |
construct | +0.109 |
deserve | +0.109 |
ws | +0.106 |
rebuild | +0.105 |
Neg logit contributions | |
ittens | -0.072 |
ls | -0.070 |
shrimp | -0.063 |
strawberries | -0.061 |
hungry | -0.061 |
Token | , |
Token ID | 11 |
Feature activation | -9.94e+00 |
Pos logit contributions | |
tted | +0.079 |
deserve | +0.076 |
construct | +0.074 |
rebuild | +0.073 |
make | +0.073 |
Neg logit contributions | |
ittens | -0.054 |
ls | -0.046 |
ipop | -0.041 |
ip | -0.041 |
ule | -0.039 |
Token | like |
Token ID | 588 |
Feature activation | -3.06e-01 |
Pos logit contributions | |
ittens | +0.316 |
ls | +0.303 |
ogging | +0.272 |
ip | +0.271 |
ule | +0.270 |
Neg logit contributions | |
tted | -0.235 |
deserve | -0.224 |
construct | -0.215 |
respect | -0.205 |
rebuild | -0.204 |
Token | cows |
Token ID | 22575 |
Feature activation | +8.99e-03 |
Pos logit contributions | |
ittens | +0.007 |
ls | +0.007 |
ogging | +0.006 |
ip | +0.006 |
ipop | +0.006 |
Neg logit contributions | |
tted | -0.010 |
deserve | -0.009 |
construct | -0.009 |
respect | -0.009 |
rebuild | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | -9.75e+00 |
Pos logit contributions | |
tted | +0.000 |
deserve | +0.000 |
make | +0.000 |
construct | +0.000 |
rebuild | +0.000 |
Neg logit contributions | |
ittens | -0.000 |
ls | -0.000 |
ogging | -0.000 |
ipop | -0.000 |
ule | -0.000 |
Token | pigs |
Token ID | 22333 |
Feature activation | -3.09e-01 |
Pos logit contributions | |
ittens | +0.230 |
ls | +0.209 |
ogging | +0.198 |
ule | +0.188 |
ip | +0.187 |
Neg logit contributions | |
make | -0.294 |
deserve | -0.292 |
tted | -0.285 |
construct | -0.283 |
build | -0.280 |
Token | , |
Token ID | 11 |
Feature activation | -9.95e+00 |
Pos logit contributions | |
ittens | +0.005 |
ls | +0.004 |
ogging | +0.003 |
strawberries | +0.003 |
ipop | +0.003 |
Neg logit contributions | |
tted | -0.012 |
deserve | -0.012 |
construct | -0.012 |
rebuild | -0.012 |
make | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | -8.30e+00 |
Pos logit contributions | |
ittens | +0.199 |
ls | +0.178 |
ogging | +0.160 |
ule | +0.151 |
ip | +0.150 |
Neg logit contributions | |
deserve | -0.335 |
tted | -0.334 |
make | -0.329 |
construct | -0.327 |
respect | -0.324 |
Token | chickens |
Token ID | 25972 |
Feature activation | -3.45e-01 |
Pos logit contributions | |
ittens | +0.182 |
ls | +0.170 |
ogging | +0.143 |
ip | +0.142 |
ule | +0.139 |
Neg logit contributions | |
tted | -0.274 |
deserve | -0.270 |
construct | -0.264 |
respect | -0.256 |
rebuild | -0.256 |
Token | . |
Token ID | 13 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
ittens | +0.008 |
ls | +0.008 |
ogging | +0.007 |
ule | +0.006 |
strawberries | +0.006 |
Neg logit contributions | |
tted | -0.011 |
deserve | -0.010 |
construct | -0.010 |
rebuild | -0.010 |
respect | -0.010 |
Token | They |
Token ID | 1119 |
Feature activation | +3.33e+00 |
Pos logit contributions | |
tted | +0.039 |
deserve | +0.037 |
construct | +0.036 |
rebuild | +0.034 |
ws | +0.034 |
Neg logit contributions | |
ittens | -0.041 |
ls | -0.039 |
shrimp | -0.034 |
ogging | -0.034 |
ip | -0.034 |
Token | also |
Token ID | 635 |
Feature activation | +3.21e-01 |
Pos logit contributions | |
tted | +0.086 |
deserve | +0.083 |
construct | +0.081 |
rebuild | +0.077 |
respect | +0.076 |
Neg logit contributions | |
ittens | -0.099 |
ls | -0.094 |
ogging | -0.084 |
ip | -0.082 |
ule | -0.081 |
Token | was |
Token ID | 373 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
tted | +0.056 |
deserve | +0.053 |
construct | +0.052 |
ws | +0.051 |
rebuild | +0.050 |
Neg logit contributions | |
ittens | -0.047 |
ls | -0.046 |
she | -0.039 |
shrimp | -0.039 |
ip | -0.038 |
Token | full |
Token ID | 1336 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
ittens | +0.037 |
ls | +0.036 |
ogging | +0.031 |
ule | +0.031 |
ip | +0.031 |
Neg logit contributions | |
tted | -0.039 |
deserve | -0.037 |
construct | -0.037 |
rebuild | -0.035 |
respect | -0.034 |
Token | of |
Token ID | 286 |
Feature activation | +3.23e+00 |
Pos logit contributions | |
tted | +0.059 |
deserve | +0.056 |
construct | +0.054 |
rebuild | +0.052 |
respect | +0.052 |
Neg logit contributions | |
ittens | -0.049 |
ls | -0.047 |
ip | -0.039 |
ule | -0.039 |
ogging | -0.039 |
Token | colourful |
Token ID | 44799 |
Feature activation | +7.13e-02 |
Pos logit contributions | |
tted | +0.099 |
construct | +0.096 |
deserve | +0.095 |
respect | +0.093 |
rebuild | +0.091 |
Neg logit contributions | |
ittens | -0.079 |
ls | -0.076 |
ip | -0.065 |
ule | -0.064 |
ipop | -0.064 |
Token | shapes |
Token ID | 15268 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
construct | +0.002 |
tted | +0.002 |
respect | +0.002 |
make | +0.002 |
deserve | +0.002 |
Neg logit contributions | |
ittens | -0.002 |
ls | -0.002 |
ipop | -0.002 |
ule | -0.002 |
ip | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | -9.24e+00 |
Pos logit contributions | |
tted | +0.061 |
deserve | +0.059 |
construct | +0.058 |
rebuild | +0.056 |
respect | +0.055 |
Neg logit contributions | |
ittens | -0.040 |
ls | -0.035 |
ule | -0.030 |
ip | -0.030 |
ogging | -0.029 |
Token | toys |
Token ID | 14958 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
ittens | +0.265 |
ls | +0.252 |
ogging | +0.223 |
ule | +0.221 |
ip | +0.220 |
Neg logit contributions | |
tted | -0.247 |
deserve | -0.235 |
construct | -0.232 |
rebuild | -0.220 |
respect | -0.217 |
Token | ! |
Token ID | 0 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
ittens | +0.003 |
ls | +0.003 |
ule | +0.002 |
ip | +0.002 |
ogging | +0.002 |
Neg logit contributions | |
tted | -0.005 |
deserve | -0.005 |
construct | -0.005 |
rebuild | -0.005 |
ws | -0.005 |
Token | They |
Token ID | 1119 |
Feature activation | +2.93e+00 |
Pos logit contributions | |
tted | +0.005 |
deserve | +0.004 |
construct | +0.004 |
rebuild | +0.004 |
respect | +0.004 |
Neg logit contributions | |
ittens | -0.006 |
ls | -0.006 |
ule | -0.005 |
ipop | -0.005 |
ries | -0.005 |
Token | were |
Token ID | 547 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
tted | +0.080 |
deserve | +0.077 |
construct | +0.077 |
rebuild | +0.074 |
respect | +0.072 |
Neg logit contributions | |
ittens | -0.083 |
ls | -0.079 |
ogging | -0.069 |
ip | -0.068 |
ule | -0.068 |
Token | very |
Token ID | 845 |
Feature activation | -8.70e+00 |
Pos logit contributions | |
ittens | +0.024 |
ls | +0.023 |
ogging | +0.019 |
ip | +0.019 |
ipop | +0.019 |
Neg logit contributions | |
tted | -0.033 |
construct | -0.032 |
deserve | -0.032 |
rebuild | -0.031 |
overcome | -0.031 |