Token | watched |
Token ID | 7342 |
Feature activation | +9.45e-01 |
Pos logit contributions | |
maid | +0.012 |
fish | +0.011 |
Qu | +0.010 |
ora | +0.010 |
bl | +0.010 |
Neg logit contributions | |
supposed | -0.012 |
adders | -0.012 |
neys | -0.011 |
about | -0.011 |
online | -0.010 |
Token | as |
Token ID | 355 |
Feature activation | -9.60e-01 |
Pos logit contributions | |
supposed | +0.040 |
adders | +0.037 |
neys | +0.035 |
By | +0.031 |
done | +0.031 |
Neg logit contributions | |
maid | -0.039 |
fish | -0.037 |
breath | -0.036 |
bl | -0.035 |
dolphin | -0.034 |
Token | people |
Token ID | 661 |
Feature activation | +4.76e+00 |
Pos logit contributions | |
fish | +0.040 |
maid | +0.039 |
breath | +0.037 |
dolphin | +0.036 |
bl | +0.036 |
Neg logit contributions | |
supposed | -0.038 |
adders | -0.035 |
neys | -0.032 |
By | -0.030 |
online | -0.030 |
Token | attached |
Token ID | 7223 |
Feature activation | +8.09e-01 |
Pos logit contributions | |
supposed | +0.205 |
adders | +0.202 |
neys | +0.185 |
By | +0.176 |
emergencies | +0.172 |
Neg logit contributions | |
fish | -0.182 |
maid | -0.178 |
bl | -0.165 |
Qu | -0.159 |
breath | -0.156 |
Token | suit |
Token ID | 6050 |
Feature activation | -3.70e-01 |
Pos logit contributions | |
supposed | +0.032 |
adders | +0.030 |
By | +0.027 |
neys | +0.025 |
emergencies | +0.025 |
Neg logit contributions | |
maid | -0.033 |
breath | -0.033 |
fish | -0.031 |
bl | -0.030 |
ora | -0.028 |
Token | cases |
Token ID | 33964 |
Feature activation | +8.94e+00 |
Pos logit contributions | |
maid | +0.016 |
fish | +0.016 |
bl | +0.015 |
of | +0.014 |
ora | +0.014 |
Neg logit contributions | |
supposed | -0.014 |
adders | -0.013 |
emergencies | -0.012 |
By | -0.011 |
measuring | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
adders | +0.404 |
supposed | +0.402 |
emergencies | +0.373 |
neys | +0.357 |
elev | +0.332 |
Neg logit contributions | |
fish | -0.297 |
breath | -0.296 |
maid | -0.287 |
bl | -0.271 |
of | -0.250 |
Token | a |
Token ID | 257 |
Feature activation | -7.21e-01 |
Pos logit contributions | |
maid | +0.008 |
fish | +0.007 |
bl | +0.007 |
breath | +0.007 |
dolphin | +0.007 |
Neg logit contributions | |
supposed | -0.007 |
adders | -0.007 |
neys | -0.006 |
By | -0.006 |
done | -0.006 |
Token | big |
Token ID | 1263 |
Feature activation | +3.64e-01 |
Pos logit contributions | |
maid | +0.028 |
fish | +0.027 |
bl | +0.026 |
dolphin | +0.026 |
Qu | +0.025 |
Neg logit contributions | |
adders | -0.031 |
supposed | -0.031 |
neys | -0.027 |
By | -0.026 |
about | -0.026 |
Token | belt |
Token ID | 10999 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
supposed | +0.016 |
adders | +0.016 |
neys | +0.014 |
By | +0.013 |
about | +0.013 |
Neg logit contributions | |
maid | -0.014 |
fish | -0.013 |
bl | -0.013 |
breath | -0.012 |
Qu | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +4.72e-01 |
Pos logit contributions | |
supposed | +0.052 |
adders | +0.049 |
neys | +0.045 |
emergencies | +0.044 |
By | +0.041 |
Neg logit contributions | |
fish | -0.051 |
maid | -0.048 |
bl | -0.047 |
Qu | -0.042 |
breath | -0.042 |
Token | tall |
Token ID | 7331 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
maid | +0.067 |
bl | +0.064 |
fish | +0.063 |
breath | +0.061 |
dolphin | +0.057 |
Neg logit contributions | |
supposed | -0.090 |
adders | -0.090 |
neys | -0.082 |
By | -0.081 |
online | -0.077 |
Token | towers |
Token ID | 18028 |
Feature activation | +6.75e+00 |
Pos logit contributions | |
supposed | +0.078 |
adders | +0.069 |
neys | +0.061 |
By | +0.061 |
online | +0.059 |
Neg logit contributions | |
maid | -0.079 |
fish | -0.078 |
bl | -0.076 |
lizard | -0.073 |
breath | -0.072 |
Token | and |
Token ID | 290 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
supposed | +0.302 |
adders | +0.295 |
emergencies | +0.274 |
neys | +0.255 |
By | +0.254 |
Neg logit contributions | |
fish | -0.236 |
maid | -0.234 |
bl | -0.218 |
ora | -0.208 |
breath | -0.198 |
Token | winding |
Token ID | 28967 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
supposed | +0.009 |
adders | +0.009 |
By | +0.008 |
neys | +0.008 |
online | +0.008 |
Neg logit contributions | |
maid | -0.009 |
fish | -0.009 |
bl | -0.009 |
Qu | -0.009 |
mer | -0.009 |
Token | stair |
Token ID | 18761 |
Feature activation | +2.85e+00 |
Pos logit contributions | |
supposed | +0.095 |
adders | +0.090 |
neys | +0.075 |
By | +0.075 |
online | +0.074 |
Neg logit contributions | |
maid | -0.134 |
fish | -0.130 |
ora | -0.124 |
breath | -0.123 |
bl | -0.122 |
Token | cases |
Token ID | 33964 |
Feature activation | +8.81e+00 |
Pos logit contributions | |
supposed | +0.110 |
adders | +0.100 |
emergencies | +0.093 |
OK | +0.088 |
measuring | +0.087 |
Neg logit contributions | |
fish | -0.121 |
maid | -0.112 |
bl | -0.112 |
ora | -0.109 |
breath | -0.102 |
Token | but |
Token ID | 475 |
Feature activation | +4.55e-01 |
Pos logit contributions | |
supposed | +0.380 |
adders | +0.371 |
emergencies | +0.352 |
neys | +0.331 |
By | +0.317 |
Neg logit contributions | |
fish | -0.305 |
breath | -0.293 |
maid | -0.287 |
bl | -0.284 |
mer | -0.262 |
Token | most |
Token ID | 749 |
Feature activation | +2.59e+00 |
Pos logit contributions | |
adders | +0.019 |
supposed | +0.019 |
neys | +0.017 |
By | +0.015 |
emergencies | +0.015 |
Neg logit contributions | |
fish | -0.017 |
maid | -0.016 |
bl | -0.016 |
breath | -0.015 |
Qu | -0.015 |
Token | of |
Token ID | 286 |
Feature activation | +5.04e-01 |
Pos logit contributions | |
adders | +0.089 |
supposed | +0.087 |
online | +0.081 |
By | +0.077 |
Until | +0.075 |
Neg logit contributions | |
fish | -0.112 |
maid | -0.104 |
lizard | -0.103 |
bl | -0.097 |
snail | -0.097 |
Token | all |
Token ID | 477 |
Feature activation | -9.35e-01 |
Pos logit contributions | |
supposed | +0.018 |
adders | +0.018 |
neys | +0.016 |
By | +0.016 |
online | +0.015 |
Neg logit contributions | |
maid | -0.022 |
fish | -0.021 |
bl | -0.020 |
breath | -0.020 |
Qu | -0.019 |
Token | she |
Token ID | 673 |
Feature activation | +6.35e-01 |
Pos logit contributions | |
fish | +0.031 |
maid | +0.029 |
bl | +0.028 |
breath | +0.028 |
ora | +0.027 |
Neg logit contributions | |
adders | -0.040 |
supposed | -0.038 |
neys | -0.037 |
eps | -0.035 |
tr | -0.035 |
Token | when |
Token ID | 618 |
Feature activation | +5.70e-01 |
Pos logit contributions | |
adders | +0.036 |
supposed | +0.035 |
neys | +0.033 |
eps | +0.029 |
emergencies | +0.028 |
Neg logit contributions | |
maid | -0.033 |
bl | -0.030 |
breath | -0.029 |
ora | -0.029 |
fish | -0.029 |
Token | they |
Token ID | 484 |
Feature activation | +9.29e-01 |
Pos logit contributions | |
adders | +0.025 |
supposed | +0.024 |
neys | +0.023 |
eps | +0.020 |
online | +0.019 |
Neg logit contributions | |
maid | -0.021 |
breath | -0.020 |
bl | -0.019 |
fish | -0.019 |
Qu | -0.019 |
Token | opened |
Token ID | 4721 |
Feature activation | -4.93e-02 |
Pos logit contributions | |
supposed | +0.030 |
adders | +0.029 |
done | +0.027 |
about | +0.026 |
online | +0.026 |
Neg logit contributions | |
maid | -0.045 |
fish | -0.042 |
ora | -0.040 |
Qu | -0.040 |
dolphin | -0.038 |
Token | their |
Token ID | 511 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
maid | +0.002 |
breath | +0.002 |
fish | +0.002 |
bl | +0.002 |
ora | +0.002 |
Neg logit contributions | |
supposed | -0.002 |
adders | -0.002 |
neys | -0.002 |
By | -0.002 |
emergencies | -0.002 |
Token | suit |
Token ID | 6050 |
Feature activation | -8.24e-01 |
Pos logit contributions | |
supposed | +0.008 |
adders | +0.008 |
neys | +0.007 |
By | +0.007 |
measuring | +0.007 |
Neg logit contributions | |
maid | -0.009 |
bl | -0.008 |
fish | -0.007 |
breath | -0.007 |
ora | -0.007 |
Token | cases |
Token ID | 33964 |
Feature activation | +8.79e+00 |
Pos logit contributions | |
maid | +0.038 |
fish | +0.036 |
bl | +0.036 |
ora | +0.033 |
breath | +0.032 |
Neg logit contributions | |
supposed | -0.031 |
adders | -0.029 |
neys | -0.025 |
emergencies | -0.024 |
By | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
adders | +0.371 |
supposed | +0.370 |
neys | +0.330 |
emergencies | +0.308 |
By | +0.288 |
Neg logit contributions | |
maid | -0.345 |
fish | -0.330 |
breath | -0.321 |
bl | -0.318 |
ora | -0.301 |
Token | they |
Token ID | 484 |
Feature activation | +8.48e-01 |
Pos logit contributions | |
maid | +0.001 |
fish | +0.000 |
bl | +0.000 |
dolphin | +0.000 |
breath | +0.000 |
Neg logit contributions | |
supposed | -0.001 |
adders | -0.001 |
neys | -0.001 |
opl | -0.001 |
By | -0.000 |
Token | discovered |
Token ID | 5071 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
supposed | +0.034 |
adders | +0.034 |
neys | +0.030 |
By | +0.029 |
measuring | +0.029 |
Neg logit contributions | |
maid | -0.035 |
fish | -0.032 |
Qu | -0.030 |
bl | -0.030 |
ora | -0.030 |
Token | a |
Token ID | 257 |
Feature activation | -4.64e-01 |
Pos logit contributions | |
supposed | +0.064 |
adders | +0.061 |
neys | +0.057 |
By | +0.056 |
emergencies | +0.052 |
Neg logit contributions | |
maid | -0.073 |
bl | -0.069 |
fish | -0.068 |
dolphin | -0.067 |
ora | -0.062 |
Token | fast |
Token ID | 3049 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
maid | +0.021 |
fish | +0.020 |
Qu | +0.019 |
bl | +0.019 |
ora | +0.019 |
Neg logit contributions | |
adders | -0.016 |
supposed | -0.016 |
neys | -0.014 |
By | -0.013 |
about | -0.013 |
Token | was |
Token ID | 373 |
Feature activation | +6.59e-01 |
Pos logit contributions | |
adders | +0.075 |
supposed | +0.073 |
neys | +0.066 |
emergencies | +0.064 |
elev | +0.062 |
Neg logit contributions | |
maid | -0.061 |
fish | -0.060 |
bl | -0.057 |
dolphin | -0.054 |
Qu | -0.053 |
Token | busy |
Token ID | 8179 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
adders | +0.024 |
neys | +0.021 |
chairs | +0.018 |
supposed | +0.018 |
bricks | +0.017 |
Neg logit contributions | |
bl | -0.030 |
maid | -0.029 |
breath | -0.029 |
mer | -0.026 |
groom | -0.026 |
Token | packing |
Token ID | 24157 |
Feature activation | -6.45e-01 |
Pos logit contributions | |
adders | +0.077 |
supposed | +0.070 |
neys | +0.064 |
online | +0.057 |
By | +0.056 |
Neg logit contributions | |
maid | -0.109 |
bl | -0.104 |
fish | -0.103 |
breath | -0.102 |
ora | -0.097 |
Token | the |
Token ID | 262 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
maid | +0.027 |
bl | +0.026 |
breath | +0.025 |
fish | +0.025 |
ora | +0.024 |
Neg logit contributions | |
adders | -0.025 |
supposed | -0.025 |
neys | -0.022 |
By | -0.020 |
emergencies | -0.020 |
Token | suit |
Token ID | 6050 |
Feature activation | -8.84e-01 |
Pos logit contributions | |
adders | +0.043 |
supposed | +0.041 |
neys | +0.039 |
By | +0.038 |
Kim | +0.036 |
Neg logit contributions | |
bl | -0.040 |
maid | -0.038 |
dolphin | -0.038 |
fish | -0.036 |
breath | -0.036 |
Token | cases |
Token ID | 33964 |
Feature activation | +8.74e+00 |
Pos logit contributions | |
fish | +0.035 |
bl | +0.035 |
maid | +0.033 |
breath | +0.031 |
ora | +0.029 |
Neg logit contributions | |
supposed | -0.032 |
adders | -0.032 |
emergencies | -0.030 |
neys | -0.028 |
By | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
adders | +0.397 |
supposed | +0.385 |
neys | +0.353 |
emergencies | +0.352 |
elev | +0.325 |
Neg logit contributions | |
fish | -0.290 |
breath | -0.285 |
bl | -0.282 |
maid | -0.280 |
ora | -0.238 |
Token | while |
Token ID | 981 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
bl | +0.009 |
maid | +0.008 |
fish | +0.008 |
breath | +0.008 |
dolphin | +0.008 |
Neg logit contributions | |
adders | -0.009 |
supposed | -0.009 |
neys | -0.008 |
By | -0.007 |
emergencies | -0.007 |
Token | Daddy |
Token ID | 26926 |
Feature activation | +6.45e-01 |
Pos logit contributions | |
maid | +0.089 |
bl | +0.085 |
breath | +0.082 |
dolphin | +0.080 |
fish | +0.078 |
Neg logit contributions | |
adders | -0.111 |
supposed | -0.105 |
neys | -0.103 |
case | -0.090 |
eps | -0.087 |
Token | was |
Token ID | 373 |
Feature activation | +7.08e-01 |
Pos logit contributions | |
adders | +0.029 |
supposed | +0.027 |
neys | +0.025 |
emergencies | +0.024 |
elev | +0.024 |
Neg logit contributions | |
maid | -0.023 |
fish | -0.023 |
bl | -0.022 |
Qu | -0.021 |
dolphin | -0.020 |
Token | busy |
Token ID | 8179 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
adders | +0.022 |
neys | +0.020 |
supposed | +0.017 |
bricks | +0.017 |
chairs | +0.016 |
Neg logit contributions | |
bl | -0.034 |
maid | -0.034 |
breath | -0.033 |
fish | -0.032 |
Qu | -0.030 |
Token | They |
Token ID | 1119 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
Qu | +0.008 |
maid | +0.008 |
bl | +0.007 |
ora | +0.007 |
fish | +0.007 |
Neg logit contributions | |
supposed | -0.011 |
adders | -0.010 |
done | -0.009 |
neys | -0.009 |
emergencies | -0.009 |
Token | would |
Token ID | 561 |
Feature activation | -3.63e-01 |
Pos logit contributions | |
supposed | +0.006 |
adders | +0.006 |
neys | +0.005 |
emergencies | +0.005 |
By | +0.005 |
Neg logit contributions | |
maid | -0.005 |
fish | -0.005 |
bl | -0.005 |
Qu | -0.005 |
ora | -0.005 |
Token | fill |
Token ID | 6070 |
Feature activation | -2.97e+00 |
Pos logit contributions | |
maid | +0.015 |
fish | +0.014 |
Qu | +0.014 |
ora | +0.014 |
bl | +0.013 |
Neg logit contributions | |
supposed | -0.014 |
adders | -0.014 |
neys | -0.012 |
about | -0.012 |
online | -0.012 |
Token | their |
Token ID | 511 |
Feature activation | +3.39e-02 |
Pos logit contributions | |
maid | +0.135 |
fish | +0.126 |
bl | +0.122 |
Qu | +0.120 |
breath | +0.120 |
Neg logit contributions | |
supposed | -0.109 |
adders | -0.106 |
neys | -0.090 |
By | -0.090 |
emergencies | -0.085 |
Token | suit |
Token ID | 6050 |
Feature activation | -9.57e-01 |
Pos logit contributions | |
supposed | +0.001 |
adders | +0.001 |
By | +0.001 |
about | +0.001 |
neys | +0.001 |
Neg logit contributions | |
maid | -0.001 |
bl | -0.001 |
breath | -0.001 |
fish | -0.001 |
Qu | -0.001 |
Token | cases |
Token ID | 33964 |
Feature activation | +8.70e+00 |
Pos logit contributions | |
maid | +0.041 |
fish | +0.040 |
bl | +0.038 |
breath | +0.037 |
ora | +0.036 |
Neg logit contributions | |
supposed | -0.037 |
adders | -0.035 |
emergencies | -0.033 |
neys | -0.030 |
By | -0.030 |
Token | with |
Token ID | 351 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
supposed | +0.376 |
adders | +0.368 |
emergencies | +0.322 |
neys | +0.318 |
By | +0.301 |
Neg logit contributions | |
maid | -0.335 |
fish | -0.318 |
breath | -0.307 |
bl | -0.299 |
ora | -0.290 |
Token | clothes |
Token ID | 8242 |
Feature activation | +9.26e-01 |
Pos logit contributions | |
bl | +0.078 |
maid | +0.077 |
breath | +0.075 |
fish | +0.072 |
dolphin | +0.071 |
Neg logit contributions | |
supposed | -0.088 |
adders | -0.085 |
By | -0.081 |
neys | -0.077 |
Until | -0.071 |
Token | and |
Token ID | 290 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
supposed | +0.042 |
adders | +0.041 |
emergencies | +0.037 |
neys | +0.036 |
By | +0.036 |
Neg logit contributions | |
maid | -0.034 |
fish | -0.033 |
bl | -0.031 |
breath | -0.029 |
ora | -0.028 |
Token | toys |
Token ID | 14958 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
supposed | +0.008 |
adders | +0.008 |
By | +0.008 |
neys | +0.008 |
about | +0.007 |
Neg logit contributions | |
bl | -0.005 |
maid | -0.005 |
breath | -0.005 |
fish | -0.005 |
dolphin | -0.005 |
Token | for |
Token ID | 329 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
supposed | +0.111 |
adders | +0.105 |
emergencies | +0.100 |
By | +0.098 |
Grad | +0.094 |
Neg logit contributions | |
fish | -0.096 |
maid | -0.087 |
bl | -0.085 |
breath | -0.079 |
ora | -0.075 |
Token | different |
Token ID | 1180 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
supposed | +0.020 |
adders | +0.020 |
By | +0.018 |
neys | +0.018 |
Until | +0.017 |
Neg logit contributions | |
maid | -0.019 |
bl | -0.019 |
fish | -0.018 |
dolphin | -0.018 |
breath | -0.018 |
Token | people |
Token ID | 661 |
Feature activation | +4.04e+00 |
Pos logit contributions | |
adders | +0.065 |
supposed | +0.064 |
By | +0.056 |
online | +0.055 |
measuring | +0.055 |
Neg logit contributions | |
maid | -0.053 |
fish | -0.048 |
bl | -0.046 |
breath | -0.046 |
ora | -0.045 |
Token | with |
Token ID | 351 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
adders | +0.180 |
Until | +0.160 |
By | +0.157 |
neys | +0.155 |
Hours | +0.152 |
Neg logit contributions | |
fish | -0.160 |
maid | -0.150 |
bl | -0.146 |
dolphin | -0.129 |
ora | -0.124 |
Token | huge |
Token ID | 3236 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
maid | +0.075 |
bl | +0.071 |
fish | +0.070 |
breath | +0.070 |
Qu | +0.067 |
Neg logit contributions | |
supposed | -0.077 |
adders | -0.076 |
neys | -0.069 |
By | -0.068 |
online | -0.063 |
Token | suit |
Token ID | 6050 |
Feature activation | -8.93e-01 |
Pos logit contributions | |
supposed | +0.041 |
adders | +0.039 |
neys | +0.035 |
By | +0.035 |
online | +0.033 |
Neg logit contributions | |
maid | -0.041 |
bl | -0.038 |
breath | -0.037 |
fish | -0.037 |
dolphin | -0.035 |
Token | cases |
Token ID | 33964 |
Feature activation | +8.61e+00 |
Pos logit contributions | |
maid | +0.040 |
fish | +0.040 |
bl | +0.039 |
ora | +0.035 |
dolphin | +0.034 |
Neg logit contributions | |
adders | -0.029 |
supposed | -0.029 |
By | -0.027 |
emergencies | -0.026 |
Grad | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -7.76e-02 |
Pos logit contributions | |
adders | +0.391 |
supposed | +0.363 |
neys | +0.360 |
emergencies | +0.338 |
By | +0.336 |
Neg logit contributions | |
maid | -0.281 |
bl | -0.278 |
fish | -0.275 |
breath | -0.263 |
mer | -0.229 |
Token | She |
Token ID | 1375 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
maid | +0.003 |
ora | +0.003 |
Qu | +0.003 |
fish | +0.003 |
bl | +0.003 |
Neg logit contributions | |
supposed | -0.003 |
adders | -0.003 |
neys | -0.002 |
measuring | -0.002 |
done | -0.002 |
Token | wanted |
Token ID | 2227 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
maid | +0.007 |
fish | +0.006 |
ora | +0.006 |
Qu | +0.006 |
breath | +0.006 |
Neg logit contributions | |
supposed | -0.006 |
adders | -0.006 |
about | -0.005 |
neys | -0.005 |
driving | -0.005 |
Token | to |
Token ID | 284 |
Feature activation | -5.69e-01 |
Pos logit contributions | |
adders | +0.054 |
supposed | +0.050 |
neys | +0.047 |
By | +0.044 |
Until | +0.043 |
Neg logit contributions | |
bl | -0.055 |
maid | -0.053 |
dolphin | -0.051 |
fish | -0.051 |
mer | -0.048 |
Token | look |
Token ID | 804 |
Feature activation | -2.95e+00 |
Pos logit contributions | |
maid | +0.025 |
fish | +0.023 |
Qu | +0.022 |
ora | +0.022 |
bl | +0.022 |
Neg logit contributions | |
supposed | -0.021 |
adders | -0.021 |
neys | -0.018 |
about | -0.018 |
online | -0.018 |
Token | to |
Token ID | 284 |
Feature activation | -7.37e-01 |
Pos logit contributions | |
adders | +0.022 |
neys | +0.020 |
supposed | +0.020 |
chairs | +0.019 |
elev | +0.018 |
Neg logit contributions | |
breath | -0.023 |
fish | -0.023 |
maid | -0.022 |
bl | -0.021 |
dolphin | -0.021 |
Token | go |
Token ID | 467 |
Feature activation | +5.10e-01 |
Pos logit contributions | |
maid | +0.029 |
fish | +0.028 |
Qu | +0.027 |
bl | +0.027 |
breath | +0.026 |
Neg logit contributions | |
adders | -0.029 |
supposed | -0.029 |
neys | -0.026 |
online | -0.025 |
By | -0.024 |
Token | with |
Token ID | 351 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
adders | +0.017 |
supposed | +0.016 |
neys | +0.015 |
By | +0.013 |
emergencies | +0.013 |
Neg logit contributions | |
maid | -0.023 |
fish | -0.023 |
bl | -0.022 |
breath | -0.022 |
ora | -0.022 |
Token | their |
Token ID | 511 |
Feature activation | +5.31e-02 |
Pos logit contributions | |
maid | +0.095 |
bl | +0.091 |
breath | +0.090 |
dolphin | +0.088 |
fish | +0.088 |
Neg logit contributions | |
supposed | -0.080 |
adders | -0.079 |
neys | -0.070 |
By | -0.069 |
about | -0.064 |
Token | suit |
Token ID | 6050 |
Feature activation | -9.73e-01 |
Pos logit contributions | |
adders | +0.002 |
supposed | +0.002 |
neys | +0.002 |
By | +0.002 |
about | +0.002 |
Neg logit contributions | |
maid | -0.002 |
bl | -0.002 |
breath | -0.002 |
dolphin | -0.002 |
fish | -0.002 |
Token | cases |
Token ID | 33964 |
Feature activation | +8.57e+00 |
Pos logit contributions | |
bl | +0.045 |
maid | +0.044 |
fish | +0.044 |
breath | +0.041 |
ora | +0.040 |
Neg logit contributions | |
adders | -0.029 |
supposed | -0.029 |
emergencies | -0.026 |
By | -0.026 |
neys | -0.024 |
Token | packed |
Token ID | 11856 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
adders | +0.401 |
supposed | +0.377 |
neys | +0.356 |
emergencies | +0.346 |
elev | +0.330 |
Neg logit contributions | |
fish | -0.271 |
maid | -0.270 |
bl | -0.267 |
breath | -0.266 |
ora | -0.229 |
Token | , |
Token ID | 11 |
Feature activation | -3.37e-01 |
Pos logit contributions | |
maid | +0.118 |
fish | +0.112 |
bl | +0.106 |
Qu | +0.106 |
ora | +0.106 |
Neg logit contributions | |
adders | -0.084 |
supposed | -0.083 |
neys | -0.071 |
By | -0.069 |
emergencies | -0.067 |
Token | but |
Token ID | 475 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
bl | +0.013 |
maid | +0.012 |
breath | +0.012 |
fish | +0.012 |
dolphin | +0.012 |
Neg logit contributions | |
adders | -0.013 |
supposed | -0.013 |
neys | -0.012 |
By | -0.011 |
emergencies | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | +8.76e-01 |
Pos logit contributions | |
supposed | +0.006 |
adders | +0.006 |
neys | +0.005 |
By | +0.005 |
emergencies | +0.005 |
Neg logit contributions | |
maid | -0.006 |
fish | -0.006 |
bl | -0.006 |
breath | -0.005 |
Qu | -0.005 |
Token | person |
Token ID | 1048 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
supposed | +0.034 |
adders | +0.034 |
neys | +0.029 |
about | +0.028 |
online | +0.028 |
Neg logit contributions | |
maid | -0.037 |
fish | -0.035 |
Qu | -0.033 |
ora | -0.033 |
bl | -0.032 |
Token | He |
Token ID | 679 |
Feature activation | -3.00e-01 |
Pos logit contributions | |
supposed | +0.004 |
adders | +0.004 |
neys | +0.003 |
done | +0.003 |
online | +0.003 |
Neg logit contributions | |
Qu | -0.004 |
maid | -0.004 |
bl | -0.003 |
dolphin | -0.003 |
breath | -0.003 |
Token | put |
Token ID | 1234 |
Feature activation | -7.59e-01 |
Pos logit contributions | |
maid | +0.011 |
fish | +0.011 |
bl | +0.010 |
Qu | +0.010 |
ora | +0.010 |
Neg logit contributions | |
adders | -0.013 |
supposed | -0.013 |
neys | -0.011 |
By | -0.011 |
emergencies | -0.011 |
Token | all |
Token ID | 477 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
maid | +0.032 |
bl | +0.032 |
breath | +0.031 |
fish | +0.030 |
Qu | +0.029 |
Neg logit contributions | |
supposed | -0.029 |
adders | -0.027 |
neys | -0.025 |
By | -0.024 |
emergencies | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | +9.32e-01 |
Pos logit contributions | |
bl | +0.038 |
breath | +0.038 |
maid | +0.034 |
fish | +0.033 |
Qu | +0.032 |
Neg logit contributions | |
adders | -0.046 |
supposed | -0.045 |
neys | -0.040 |
emergencies | -0.038 |
By | -0.038 |
Token | suit |
Token ID | 6050 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
adders | +0.036 |
supposed | +0.036 |
By | +0.032 |
neys | +0.032 |
about | +0.031 |
Neg logit contributions | |
maid | -0.038 |
bl | -0.037 |
fish | -0.035 |
breath | -0.035 |
dolphin | -0.034 |
Token | cases |
Token ID | 33964 |
Feature activation | +8.56e+00 |
Pos logit contributions | |
bl | +0.046 |
fish | +0.044 |
maid | +0.042 |
breath | +0.042 |
of | +0.036 |
Neg logit contributions | |
supposed | -0.037 |
emergencies | -0.033 |
adders | -0.033 |
By | -0.032 |
neys | -0.030 |
Token | in |
Token ID | 287 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
supposed | +0.371 |
adders | +0.355 |
neys | +0.331 |
emergencies | +0.325 |
By | +0.305 |
Neg logit contributions | |
breath | -0.320 |
bl | -0.298 |
fish | -0.289 |
maid | -0.284 |
mer | -0.249 |
Token | the |
Token ID | 262 |
Feature activation | +9.06e-01 |
Pos logit contributions | |
maid | +0.054 |
breath | +0.051 |
bl | +0.049 |
fish | +0.049 |
Qu | +0.048 |
Neg logit contributions | |
adders | -0.043 |
supposed | -0.042 |
neys | -0.036 |
By | -0.033 |
about | -0.033 |
Token | car |
Token ID | 1097 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
supposed | +0.038 |
adders | +0.038 |
neys | +0.033 |
about | +0.032 |
By | +0.031 |
Neg logit contributions | |
bl | -0.034 |
maid | -0.034 |
Qu | -0.031 |
dolphin | -0.031 |
fish | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
supposed | +0.049 |
adders | +0.047 |
emergencies | +0.043 |
By | +0.041 |
neys | +0.040 |
Neg logit contributions | |
bl | -0.035 |
fish | -0.035 |
maid | -0.033 |
breath | -0.031 |
dolphin | -0.029 |
Token | and |
Token ID | 290 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
bl | +0.012 |
maid | +0.012 |
fish | +0.011 |
breath | +0.011 |
dolphin | +0.011 |
Neg logit contributions | |
supposed | -0.013 |
adders | -0.013 |
neys | -0.012 |
By | -0.011 |
emergencies | -0.010 |
Token | running |
Token ID | 2491 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
adders | +0.168 |
supposed | +0.167 |
neys | +0.151 |
By | +0.149 |
emergencies | +0.147 |
Neg logit contributions | |
fish | -0.156 |
maid | -0.154 |
bl | -0.144 |
breath | -0.133 |
ora | -0.133 |
Token | around |
Token ID | 1088 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
supposed | +0.040 |
adders | +0.038 |
neys | +0.033 |
emergencies | +0.032 |
measuring | +0.032 |
Neg logit contributions | |
maid | -0.050 |
fish | -0.049 |
breath | -0.048 |
bl | -0.047 |
ora | -0.047 |
Token | with |
Token ID | 351 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
supposed | +0.051 |
adders | +0.050 |
neys | +0.044 |
By | +0.041 |
online | +0.040 |
Neg logit contributions | |
maid | -0.059 |
fish | -0.056 |
bl | -0.054 |
breath | -0.054 |
Qu | -0.053 |
Token | their |
Token ID | 511 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
maid | +0.068 |
fish | +0.063 |
bl | +0.062 |
breath | +0.062 |
Qu | +0.061 |
Neg logit contributions | |
supposed | -0.062 |
adders | -0.060 |
neys | -0.054 |
By | -0.052 |
online | -0.050 |
Token | suit |
Token ID | 6050 |
Feature activation | -8.18e-01 |
Pos logit contributions | |
supposed | +0.006 |
adders | +0.005 |
By | +0.005 |
neys | +0.005 |
about | +0.005 |
Neg logit contributions | |
breath | -0.005 |
maid | -0.005 |
bl | -0.005 |
fish | -0.005 |
dolphin | -0.004 |
Token | cases |
Token ID | 33964 |
Feature activation | +8.55e+00 |
Pos logit contributions | |
maid | +0.032 |
bl | +0.031 |
fish | +0.031 |
breath | +0.028 |
ora | +0.027 |
Neg logit contributions | |
supposed | -0.033 |
adders | -0.032 |
emergencies | -0.029 |
By | -0.028 |
neys | -0.027 |
Token | looked |
Token ID | 3114 |
Feature activation | -6.67e-01 |
Pos logit contributions | |
adders | +0.360 |
supposed | +0.350 |
neys | +0.325 |
emergencies | +0.303 |
By | +0.289 |
Neg logit contributions | |
maid | -0.330 |
fish | -0.322 |
bl | -0.318 |
breath | -0.309 |
dolphin | -0.281 |
Token | like |
Token ID | 588 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
maid | +0.031 |
bl | +0.029 |
fish | +0.029 |
breath | +0.027 |
dolphin | +0.027 |
Neg logit contributions | |
adders | -0.025 |
supposed | -0.024 |
neys | -0.022 |
By | -0.020 |
emergencies | -0.020 |
Token | tiny |
Token ID | 7009 |
Feature activation | -7.56e-01 |
Pos logit contributions | |
supposed | +0.075 |
adders | +0.072 |
neys | +0.066 |
By | +0.063 |
online | +0.057 |
Neg logit contributions | |
maid | -0.090 |
bl | -0.088 |
fish | -0.087 |
dolphin | -0.086 |
mer | -0.083 |
Token | ants |
Token ID | 27842 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
maid | +0.031 |
bl | +0.029 |
fish | +0.028 |
breath | +0.027 |
dolphin | +0.027 |
Neg logit contributions | |
supposed | -0.032 |
adders | -0.031 |
By | -0.027 |
neys | -0.027 |
done | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | +4.37e-01 |
Pos logit contributions | |
adders | +0.092 |
supposed | +0.088 |
neys | +0.081 |
By | +0.077 |
emergencies | +0.075 |
Neg logit contributions | |
fish | -0.097 |
maid | -0.093 |
bl | -0.092 |
dolphin | -0.084 |
snail | -0.083 |
Token | and |
Token ID | 290 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
supposed | +0.132 |
adders | +0.127 |
neys | +0.120 |
emergencies | +0.112 |
By | +0.109 |
Neg logit contributions | |
fish | -0.154 |
maid | -0.148 |
bl | -0.141 |
breath | -0.136 |
dolphin | -0.130 |
Token | all |
Token ID | 477 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
adders | +0.006 |
supposed | +0.006 |
By | +0.005 |
neys | +0.005 |
online | +0.005 |
Neg logit contributions | |
maid | -0.005 |
fish | -0.005 |
bl | -0.005 |
breath | -0.005 |
dolphin | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
fish | +0.040 |
bl | +0.040 |
maid | +0.039 |
breath | +0.039 |
dolphin | +0.037 |
Neg logit contributions | |
adders | -0.044 |
supposed | -0.042 |
neys | -0.038 |
emergencies | -0.037 |
By | -0.036 |
Token | big |
Token ID | 1263 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
adders | +0.039 |
supposed | +0.036 |
By | +0.034 |
neys | +0.033 |
online | +0.031 |
Neg logit contributions | |
bl | -0.039 |
maid | -0.038 |
breath | -0.038 |
dolphin | -0.037 |
fish | -0.037 |
Token | suit |
Token ID | 6050 |
Feature activation | -9.58e-01 |
Pos logit contributions | |
supposed | +0.007 |
adders | +0.007 |
By | +0.006 |
neys | +0.006 |
online | +0.006 |
Neg logit contributions | |
maid | -0.007 |
fish | -0.006 |
bl | -0.006 |
breath | -0.006 |
dolphin | -0.006 |
Token | cases |
Token ID | 33964 |
Feature activation | +8.54e+00 |
Pos logit contributions | |
fish | +0.038 |
maid | +0.038 |
bl | +0.037 |
breath | +0.033 |
Qu | +0.031 |
Neg logit contributions | |
supposed | -0.037 |
adders | -0.037 |
By | -0.033 |
emergencies | -0.033 |
neys | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | +6.13e-02 |
Pos logit contributions | |
adders | +0.349 |
supposed | +0.347 |
neys | +0.321 |
emergencies | +0.295 |
By | +0.291 |
Neg logit contributions | |
fish | -0.321 |
maid | -0.320 |
bl | -0.312 |
breath | -0.310 |
Qu | -0.276 |
Token | |
Token ID | 220 |
Feature activation | -8.65e-01 |
Pos logit contributions | |
supposed | +0.003 |
adders | +0.003 |
eps | +0.002 |
neys | +0.002 |
elev | +0.002 |
Neg logit contributions | |
maid | -0.002 |
Qu | -0.002 |
ora | -0.002 |
dolphin | -0.002 |
fish | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +5.86e-01 |
Pos logit contributions | |
maid | +0.022 |
ora | +0.021 |
Qu | +0.021 |
fish | +0.020 |
of | +0.019 |
Neg logit contributions | |
supposed | -0.040 |
adders | -0.036 |
elev | -0.036 |
measuring | -0.035 |
firefighters | -0.035 |
Token |
|
Token ID | 198 |
Feature activation | +6.07e-01 |
Pos logit contributions | |
supposed | +0.023 |
adders | +0.023 |
neys | +0.020 |
emergencies | +0.019 |
elev | +0.019 |
Neg logit contributions | |
Qu | -0.023 |
fish | -0.022 |
maid | -0.022 |
ora | -0.021 |
N | -0.021 |
Token | But |
Token ID | 1537 |
Feature activation | +3.97e-01 |
Pos logit contributions | |
supposed | +0.028 |
adders | +0.028 |
neys | +0.025 |
driving | +0.023 |
Wil | +0.023 |
Neg logit contributions | |
Qu | -0.020 |
fish | -0.020 |
maid | -0.019 |
ora | -0.018 |
dolphin | -0.018 |
Token | big |
Token ID | 1263 |
Feature activation | +3.64e-01 |
Pos logit contributions | |
maid | +0.028 |
fish | +0.027 |
bl | +0.026 |
dolphin | +0.026 |
Qu | +0.025 |
Neg logit contributions | |
adders | -0.031 |
supposed | -0.031 |
neys | -0.027 |
By | -0.026 |
about | -0.026 |
Token | belt |
Token ID | 10999 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
supposed | +0.016 |
adders | +0.016 |
neys | +0.014 |
By | +0.013 |
about | +0.013 |
Neg logit contributions | |
maid | -0.014 |
fish | -0.013 |
bl | -0.013 |
breath | -0.012 |
Qu | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +4.72e-01 |
Pos logit contributions | |
supposed | +0.052 |
adders | +0.049 |
neys | +0.045 |
emergencies | +0.044 |
By | +0.041 |
Neg logit contributions | |
fish | -0.051 |
maid | -0.048 |
bl | -0.047 |
Qu | -0.042 |
breath | -0.042 |
Token | The |
Token ID | 383 |
Feature activation | +8.74e-01 |
Pos logit contributions | |
supposed | +0.021 |
adders | +0.018 |
done | +0.017 |
neys | +0.016 |
eps | +0.016 |
Neg logit contributions | |
Qu | -0.018 |
maid | -0.017 |
ora | -0.017 |
dolphin | -0.016 |
Holly | -0.016 |
Token | suit |
Token ID | 6050 |
Feature activation | -8.79e-01 |
Pos logit contributions | |
supposed | +0.037 |
adders | +0.036 |
neys | +0.032 |
By | +0.031 |
about | +0.031 |
Neg logit contributions | |
maid | -0.033 |
fish | -0.031 |
bl | -0.031 |
Qu | -0.030 |
dolphin | -0.030 |
Token | cases |
Token ID | 33964 |
Feature activation | +8.54e+00 |
Pos logit contributions | |
maid | +0.031 |
fish | +0.030 |
bl | +0.029 |
Qu | +0.027 |
breath | +0.026 |
Neg logit contributions | |
supposed | -0.041 |
adders | -0.040 |
emergencies | -0.035 |
By | -0.035 |
neys | -0.035 |
Token | went |
Token ID | 1816 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
supposed | +0.393 |
adders | +0.383 |
neys | +0.347 |
emergencies | +0.333 |
By | +0.323 |
Neg logit contributions | |
fish | -0.295 |
maid | -0.295 |
bl | -0.281 |
breath | -0.275 |
Qu | -0.253 |
Token | into |
Token ID | 656 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
adders | +0.051 |
supposed | +0.049 |
neys | +0.044 |
emergencies | +0.040 |
By | +0.039 |
Neg logit contributions | |
maid | -0.058 |
bl | -0.056 |
breath | -0.055 |
fish | -0.054 |
Qu | -0.054 |
Token | the |
Token ID | 262 |
Feature activation | +5.98e-01 |
Pos logit contributions | |
maid | +0.006 |
breath | +0.005 |
fish | +0.005 |
Qu | +0.005 |
bl | +0.005 |
Neg logit contributions | |
supposed | -0.004 |
adders | -0.004 |
neys | -0.003 |
online | -0.003 |
about | -0.003 |
Token | plane |
Token ID | 6614 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
adders | +0.026 |
supposed | +0.025 |
about | +0.021 |
neys | +0.021 |
By | +0.020 |
Neg logit contributions | |
bl | -0.023 |
maid | -0.023 |
dolphin | -0.022 |
breath | -0.022 |
Qu | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
supposed | +0.073 |
adders | +0.073 |
emergencies | +0.063 |
neys | +0.060 |
By | +0.058 |
Neg logit contributions | |
fish | -0.067 |
maid | -0.067 |
breath | -0.061 |
bl | -0.061 |
ora | -0.058 |
Token | to |
Token ID | 284 |
Feature activation | -4.79e-01 |
Pos logit contributions | |
adders | +0.050 |
supposed | +0.047 |
neys | +0.046 |
emergencies | +0.040 |
clocks | +0.038 |
Neg logit contributions | |
breath | -0.057 |
bl | -0.055 |
maid | -0.054 |
ora | -0.052 |
fish | -0.051 |
Token | pick |
Token ID | 2298 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
breath | +0.019 |
maid | +0.019 |
bl | +0.018 |
dolphin | +0.017 |
ora | +0.017 |
Neg logit contributions | |
adders | -0.019 |
supposed | -0.018 |
neys | -0.018 |
chairs | -0.015 |
By | -0.015 |
Token | up |
Token ID | 510 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
maid | +0.101 |
bl | +0.099 |
ora | +0.097 |
dolphin | +0.094 |
fish | +0.092 |
Neg logit contributions | |
supposed | -0.086 |
adders | -0.079 |
neys | -0.074 |
By | -0.069 |
done | -0.068 |
Token | their |
Token ID | 511 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
supposed | +0.077 |
adders | +0.071 |
neys | +0.063 |
By | +0.062 |
done | +0.059 |
Neg logit contributions | |
maid | -0.084 |
bl | -0.079 |
breath | -0.078 |
fish | -0.077 |
ora | -0.077 |
Token | suit |
Token ID | 6050 |
Feature activation | -8.41e-01 |
Pos logit contributions | |
adders | +0.010 |
supposed | +0.010 |
By | +0.009 |
neys | +0.008 |
about | +0.008 |
Neg logit contributions | |
bl | -0.009 |
maid | -0.009 |
breath | -0.009 |
fish | -0.008 |
dolphin | -0.008 |
Token | cases |
Token ID | 33964 |
Feature activation | +8.54e+00 |
Pos logit contributions | |
maid | +0.029 |
bl | +0.029 |
fish | +0.028 |
breath | +0.025 |
ora | +0.024 |
Neg logit contributions | |
supposed | -0.038 |
adders | -0.037 |
emergencies | -0.035 |
neys | -0.033 |
By | -0.032 |
Token | but |
Token ID | 475 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
adders | +0.405 |
supposed | +0.394 |
neys | +0.369 |
emergencies | +0.363 |
elev | +0.324 |
Neg logit contributions | |
breath | -0.270 |
maid | -0.270 |
fish | -0.268 |
bl | -0.262 |
ora | -0.232 |
Token | they |
Token ID | 484 |
Feature activation | +7.72e-01 |
Pos logit contributions | |
supposed | +0.017 |
adders | +0.016 |
neys | +0.015 |
By | +0.014 |
emergencies | +0.013 |
Neg logit contributions | |
maid | -0.017 |
bl | -0.016 |
fish | -0.015 |
Qu | -0.015 |
breath | -0.015 |
Token | couldn |
Token ID | 3521 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
supposed | +0.031 |
adders | +0.031 |
neys | +0.028 |
By | +0.027 |
emergencies | +0.027 |
Neg logit contributions | |
maid | -0.031 |
fish | -0.029 |
bl | -0.027 |
Qu | -0.027 |
ora | -0.026 |
Token | 't |
Token ID | 470 |
Feature activation | -4.03e-01 |
Pos logit contributions | |
supposed | +0.060 |
adders | +0.059 |
neys | +0.055 |
online | +0.052 |
emergencies | +0.051 |
Neg logit contributions | |
fish | -0.043 |
maid | -0.042 |
Qu | -0.042 |
bl | -0.041 |
ora | -0.039 |
Token | find |
Token ID | 1064 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
maid | +0.015 |
fish | +0.015 |
ora | +0.014 |
Qu | +0.013 |
breath | +0.013 |
Neg logit contributions | |
supposed | -0.017 |
adders | -0.017 |
neys | -0.015 |
By | -0.015 |
about | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
supposed | +0.077 |
adders | +0.074 |
neys | +0.065 |
emergencies | +0.060 |
online | +0.060 |
Neg logit contributions | |
maid | -0.094 |
fish | -0.089 |
ora | -0.084 |
bl | -0.083 |
dolphin | -0.082 |
Token | Tony |
Token ID | 8832 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
maid | +0.002 |
fish | +0.002 |
ora | +0.002 |
Qu | +0.002 |
dolphin | +0.002 |
Neg logit contributions | |
supposed | -0.001 |
adders | -0.001 |
neys | -0.001 |
online | -0.001 |
emergencies | -0.001 |
Token | grabbed |
Token ID | 13176 |
Feature activation | -9.14e-01 |
Pos logit contributions | |
fish | +0.046 |
maid | +0.045 |
bl | +0.043 |
ora | +0.041 |
dolphin | +0.041 |
Neg logit contributions | |
supposed | -0.056 |
adders | -0.056 |
neys | -0.050 |
emergencies | -0.048 |
By | -0.047 |
Token | the |
Token ID | 262 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
maid | +0.040 |
breath | +0.040 |
fish | +0.038 |
bl | +0.038 |
dolphin | +0.037 |
Neg logit contributions | |
supposed | -0.034 |
adders | -0.032 |
neys | -0.030 |
By | -0.027 |
emergencies | -0.026 |
Token | suit |
Token ID | 6050 |
Feature activation | -9.65e-01 |
Pos logit contributions | |
supposed | +0.045 |
adders | +0.043 |
By | +0.040 |
neys | +0.040 |
about | +0.039 |
Neg logit contributions | |
maid | -0.038 |
fish | -0.037 |
dolphin | -0.036 |
bl | -0.035 |
breath | -0.034 |
Token | cases |
Token ID | 33964 |
Feature activation | +8.52e+00 |
Pos logit contributions | |
bl | +0.033 |
fish | +0.033 |
maid | +0.032 |
breath | +0.030 |
ora | +0.027 |
Neg logit contributions | |
supposed | -0.044 |
adders | -0.042 |
emergencies | -0.040 |
neys | -0.038 |
By | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | +2.35e-01 |
Pos logit contributions | |
adders | +0.412 |
supposed | +0.410 |
emergencies | +0.375 |
neys | +0.373 |
elev | +0.339 |
Neg logit contributions | |
breath | -0.271 |
fish | -0.258 |
bl | -0.253 |
maid | -0.253 |
dolphin | -0.212 |
Token | carried |
Token ID | 5281 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
supposed | +0.011 |
adders | +0.010 |
neys | +0.009 |
By | +0.009 |
emergencies | +0.009 |
Neg logit contributions | |
maid | -0.009 |
fish | -0.008 |
bl | -0.008 |
breath | -0.008 |
dolphin | -0.007 |
Token | them |
Token ID | 606 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
maid | +0.049 |
fish | +0.045 |
breath | +0.045 |
Qu | +0.042 |
bl | +0.042 |
Neg logit contributions | |
supposed | -0.072 |
adders | -0.069 |
neys | -0.065 |
By | -0.061 |
emergencies | -0.061 |
Token | out |
Token ID | 503 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
adders | +0.045 |
supposed | +0.044 |
neys | +0.040 |
emergencies | +0.040 |
By | +0.039 |
Neg logit contributions | |
breath | -0.052 |
maid | -0.050 |
fish | -0.049 |
bl | -0.048 |
dolphin | -0.043 |
Token | of |
Token ID | 286 |
Feature activation | +4.15e-01 |
Pos logit contributions | |
supposed | +0.068 |
adders | +0.065 |
neys | +0.057 |
emergencies | +0.055 |
By | +0.053 |
Neg logit contributions | |
breath | -0.068 |
maid | -0.066 |
fish | -0.064 |
bl | -0.061 |
mer | -0.056 |
Token | and |
Token ID | 290 |
Feature activation | +4.08e-01 |
Pos logit contributions | |
maid | +0.002 |
fish | +0.002 |
Qu | +0.002 |
ora | +0.002 |
bl | +0.002 |
Neg logit contributions | |
supposed | -0.002 |
adders | -0.002 |
neys | -0.001 |
about | -0.001 |
online | -0.001 |
Token | picked |
Token ID | 6497 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
supposed | +0.015 |
adders | +0.015 |
neys | +0.013 |
online | +0.013 |
By | +0.013 |
Neg logit contributions | |
maid | -0.018 |
fish | -0.017 |
Qu | -0.016 |
ora | -0.016 |
bl | -0.016 |
Token | up |
Token ID | 510 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
maid | +0.071 |
bl | +0.065 |
fish | +0.064 |
ora | +0.064 |
dolphin | +0.063 |
Neg logit contributions | |
supposed | -0.050 |
adders | -0.050 |
neys | -0.045 |
emergencies | -0.040 |
By | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | +6.95e-01 |
Pos logit contributions | |
supposed | +0.078 |
adders | +0.073 |
neys | +0.063 |
By | +0.060 |
done | +0.058 |
Neg logit contributions | |
maid | -0.094 |
bl | -0.087 |
fish | -0.087 |
breath | -0.085 |
Qu | -0.084 |
Token | suit |
Token ID | 6050 |
Feature activation | -7.93e-01 |
Pos logit contributions | |
supposed | +0.033 |
adders | +0.032 |
By | +0.030 |
neys | +0.029 |
about | +0.028 |
Neg logit contributions | |
maid | -0.024 |
fish | -0.023 |
bl | -0.022 |
breath | -0.021 |
dolphin | -0.021 |
Token | cases |
Token ID | 33964 |
Feature activation | +8.44e+00 |
Pos logit contributions | |
maid | +0.031 |
bl | +0.030 |
fish | +0.030 |
breath | +0.028 |
ora | +0.026 |
Neg logit contributions | |
supposed | -0.033 |
adders | -0.032 |
emergencies | -0.030 |
neys | -0.029 |
By | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
supposed | +0.359 |
adders | +0.355 |
neys | +0.330 |
emergencies | +0.317 |
By | +0.291 |
Neg logit contributions | |
maid | -0.316 |
fish | -0.309 |
breath | -0.306 |
bl | -0.306 |
ora | -0.268 |
Token | They |
Token ID | 1119 |
Feature activation | -4.49e-01 |
Pos logit contributions | |
supposed | +0.015 |
adders | +0.014 |
neys | +0.012 |
done | +0.012 |
eps | +0.011 |
Neg logit contributions | |
maid | -0.015 |
bl | -0.013 |
Qu | -0.013 |
dolphin | -0.013 |
ora | -0.013 |
Token | said |
Token ID | 531 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
maid | +0.019 |
fish | +0.018 |
Qu | +0.017 |
ora | +0.017 |
bl | +0.017 |
Neg logit contributions | |
supposed | -0.017 |
adders | -0.017 |
neys | -0.015 |
By | -0.014 |
about | -0.014 |
Token | goodbye |
Token ID | 24829 |
Feature activation | -3.08e+00 |
Pos logit contributions | |
supposed | +0.108 |
adders | +0.101 |
neys | +0.097 |
emergencies | +0.089 |
online | +0.086 |
Neg logit contributions | |
maid | -0.100 |
fish | -0.093 |
bl | -0.089 |
Qu | -0.087 |
dolphin | -0.085 |
Token | to |
Token ID | 284 |
Feature activation | -5.69e-01 |
Pos logit contributions | |
maid | +0.127 |
breath | +0.119 |
fish | +0.119 |
bl | +0.116 |
ora | +0.111 |
Neg logit contributions | |
adders | -0.126 |
supposed | -0.123 |
emergencies | -0.109 |
neys | -0.105 |
elev | -0.102 |
Token | , |
Token ID | 11 |
Feature activation | -8.28e-02 |
Pos logit contributions | |
supposed | +0.002 |
adders | +0.002 |
neys | +0.002 |
emergencies | +0.001 |
eps | +0.001 |
Neg logit contributions | |
maid | -0.002 |
fish | -0.001 |
bl | -0.001 |
ora | -0.001 |
breath | -0.001 |
Token | they |
Token ID | 484 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
maid | +0.003 |
dolphin | +0.003 |
bl | +0.003 |
fish | +0.003 |
breath | +0.003 |
Neg logit contributions | |
supposed | -0.004 |
adders | -0.004 |
neys | -0.003 |
opl | -0.003 |
emergencies | -0.003 |
Token | checked |
Token ID | 10667 |
Feature activation | +4.79e-01 |
Pos logit contributions | |
supposed | +0.019 |
adders | +0.018 |
neys | +0.016 |
By | +0.016 |
emergencies | +0.016 |
Neg logit contributions | |
maid | -0.018 |
fish | -0.017 |
Qu | -0.016 |
bl | -0.016 |
ora | -0.016 |
Token | their |
Token ID | 511 |
Feature activation | -4.58e-02 |
Pos logit contributions | |
supposed | +0.018 |
adders | +0.017 |
neys | +0.015 |
By | +0.015 |
measuring | +0.014 |
Neg logit contributions | |
maid | -0.021 |
breath | -0.019 |
fish | -0.019 |
dolphin | -0.018 |
bl | -0.018 |
Token | suit |
Token ID | 6050 |
Feature activation | -9.38e-01 |
Pos logit contributions | |
maid | +0.002 |
breath | +0.002 |
bl | +0.002 |
fish | +0.002 |
dolphin | +0.002 |
Neg logit contributions | |
adders | -0.002 |
supposed | -0.002 |
By | -0.002 |
neys | -0.002 |
about | -0.001 |
Token | cases |
Token ID | 33964 |
Feature activation | +8.36e+00 |
Pos logit contributions | |
maid | +0.048 |
bl | +0.047 |
fish | +0.047 |
breath | +0.044 |
ora | +0.042 |
Neg logit contributions | |
supposed | -0.028 |
adders | -0.025 |
emergencies | -0.022 |
By | -0.021 |
neys | -0.021 |
Token | in |
Token ID | 287 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
adders | +0.373 |
supposed | +0.371 |
neys | +0.340 |
emergencies | +0.326 |
By | +0.301 |
Neg logit contributions | |
maid | -0.284 |
breath | -0.284 |
fish | -0.279 |
bl | -0.274 |
ora | -0.233 |
Token | and |
Token ID | 290 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
maid | +0.049 |
breath | +0.047 |
fish | +0.045 |
bl | +0.045 |
Qu | +0.045 |
Neg logit contributions | |
adders | -0.039 |
supposed | -0.038 |
neys | -0.032 |
By | -0.030 |
about | -0.030 |
Token | walked |
Token ID | 6807 |
Feature activation | +3.98e-01 |
Pos logit contributions | |
adders | +0.010 |
supposed | +0.010 |
By | +0.009 |
neys | +0.009 |
emergencies | +0.008 |
Neg logit contributions | |
maid | -0.009 |
bl | -0.008 |
fish | -0.008 |
breath | -0.008 |
dolphin | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | -5.01e-01 |
Pos logit contributions | |
supposed | +0.016 |
adders | +0.016 |
neys | +0.014 |
emergencies | +0.013 |
By | +0.013 |
Neg logit contributions | |
maid | -0.017 |
breath | -0.016 |
bl | -0.016 |
Qu | -0.015 |
ora | -0.015 |
Token | their |
Token ID | 511 |
Feature activation | -3.92e-02 |
Pos logit contributions | |
maid | +0.022 |
breath | +0.021 |
Qu | +0.020 |
fish | +0.020 |
dolphin | +0.020 |
Neg logit contributions | |
supposed | -0.018 |
adders | -0.018 |
neys | -0.016 |
By | -0.014 |
done | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | +7.11e-01 |
Pos logit contributions | |
maid | +0.003 |
fish | +0.003 |
dolphin | +0.003 |
Qu | +0.003 |
breath | +0.003 |
Neg logit contributions | |
supposed | -0.003 |
adders | -0.003 |
neys | -0.003 |
By | -0.003 |
done | -0.003 |
Token | dad |
Token ID | 9955 |
Feature activation | +6.06e-01 |
Pos logit contributions | |
supposed | +0.035 |
adders | +0.034 |
neys | +0.031 |
By | +0.031 |
about | +0.030 |
Neg logit contributions | |
maid | -0.024 |
fish | -0.022 |
bl | -0.021 |
Qu | -0.021 |
dolphin | -0.021 |
Token | carried |
Token ID | 5281 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
supposed | +0.027 |
adders | +0.027 |
neys | +0.024 |
emergencies | +0.023 |
By | +0.023 |
Neg logit contributions | |
maid | -0.023 |
fish | -0.022 |
bl | -0.021 |
Qu | -0.020 |
ora | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | +6.82e-01 |
Pos logit contributions | |
maid | +0.056 |
breath | +0.053 |
Qu | +0.053 |
bl | +0.052 |
ora | +0.051 |
Neg logit contributions | |
supposed | -0.062 |
adders | -0.059 |
neys | -0.054 |
By | -0.050 |
emergencies | -0.049 |
Token | suit |
Token ID | 6050 |
Feature activation | -9.59e-01 |
Pos logit contributions | |
adders | +0.033 |
supposed | +0.032 |
neys | +0.030 |
By | +0.030 |
about | +0.028 |
Neg logit contributions | |
maid | -0.024 |
bl | -0.022 |
fish | -0.022 |
breath | -0.021 |
dolphin | -0.021 |
Token | cases |
Token ID | 33964 |
Feature activation | +8.35e+00 |
Pos logit contributions | |
maid | +0.039 |
fish | +0.039 |
bl | +0.038 |
breath | +0.037 |
of | +0.035 |
Neg logit contributions | |
supposed | -0.039 |
emergencies | -0.035 |
adders | -0.035 |
By | -0.032 |
neys | -0.031 |
Token | again |
Token ID | 757 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
supposed | +0.354 |
adders | +0.350 |
emergencies | +0.326 |
neys | +0.319 |
elev | +0.291 |
Neg logit contributions | |
fish | -0.308 |
breath | -0.307 |
maid | -0.302 |
bl | -0.291 |
ora | -0.262 |
Token | . |
Token ID | 13 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
fish | +0.008 |
maid | +0.008 |
ora | +0.007 |
bl | +0.007 |
breath | +0.007 |
Neg logit contributions | |
adders | -0.009 |
supposed | -0.008 |
neys | -0.007 |
emergencies | -0.007 |
By | -0.007 |
Token | They |
Token ID | 1119 |
Feature activation | -4.07e-01 |
Pos logit contributions | |
supposed | +0.009 |
adders | +0.008 |
neys | +0.007 |
done | +0.007 |
eps | +0.007 |
Neg logit contributions | |
maid | -0.010 |
Qu | -0.010 |
ora | -0.009 |
bl | -0.009 |
fish | -0.009 |
Token | had |
Token ID | 550 |
Feature activation | +9.15e-01 |
Pos logit contributions | |
maid | +0.017 |
fish | +0.016 |
Qu | +0.016 |
ora | +0.015 |
bl | +0.015 |
Neg logit contributions | |
supposed | -0.016 |
adders | -0.015 |
neys | -0.013 |
By | -0.013 |
about | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | -9.51e-01 |
Pos logit contributions | |
adders | +0.032 |
supposed | +0.028 |
neys | +0.027 |
emergencies | +0.025 |
By | +0.025 |
Neg logit contributions | |
bl | -0.040 |
Qu | -0.039 |
breath | -0.038 |
maid | -0.038 |
dolphin | -0.035 |
Token | they |
Token ID | 484 |
Feature activation | +8.05e-01 |
Pos logit contributions | |
adders | +0.025 |
supposed | +0.024 |
neys | +0.022 |
eps | +0.020 |
online | +0.019 |
Neg logit contributions | |
maid | -0.025 |
fish | -0.023 |
bl | -0.022 |
dolphin | -0.022 |
breath | -0.022 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
supposed | +0.028 |
adders | +0.027 |
done | +0.024 |
about | +0.023 |
online | +0.023 |
Neg logit contributions | |
maid | -0.037 |
fish | -0.035 |
ora | -0.034 |
Qu | -0.034 |
breath | -0.032 |
Token | her |
Token ID | 607 |
Feature activation | -7.45e-01 |
Pos logit contributions | |
adders | +0.065 |
supposed | +0.064 |
neys | +0.058 |
By | +0.055 |
driving | +0.053 |
Neg logit contributions | |
maid | -0.066 |
fish | -0.060 |
bl | -0.060 |
ora | -0.057 |
dolphin | -0.056 |
Token | ink |
Token ID | 16882 |
Feature activation | +9.84e-01 |
Pos logit contributions | |
maid | +0.033 |
fish | +0.031 |
bl | +0.030 |
dolphin | +0.028 |
breath | +0.028 |
Neg logit contributions | |
adders | -0.029 |
supposed | -0.027 |
neys | -0.025 |
By | -0.024 |
Wil | -0.023 |
Token | master |
Token ID | 4958 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
adders | +0.037 |
supposed | +0.037 |
neys | +0.032 |
By | +0.031 |
online | +0.030 |
Neg logit contributions | |
maid | -0.043 |
fish | -0.041 |
bl | -0.039 |
ora | -0.038 |
Qu | -0.037 |
Token | pieces |
Token ID | 34154 |
Feature activation | +8.12e+00 |
Pos logit contributions | |
maid | +0.089 |
fish | +0.086 |
bl | +0.083 |
breath | +0.078 |
dolphin | +0.076 |
Neg logit contributions | |
adders | -0.076 |
supposed | -0.073 |
neys | -0.069 |
By | -0.063 |
Wil | -0.062 |
Token | . |
Token ID | 13 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
adders | +0.312 |
supposed | +0.307 |
neys | +0.277 |
emergencies | +0.270 |
Wil | +0.266 |
Neg logit contributions | |
maid | -0.324 |
fish | -0.315 |
bl | -0.304 |
ora | -0.276 |
breath | -0.272 |
Token |
|
Token ID | 198 |
Feature activation | +8.33e-01 |
Pos logit contributions | |
supposed | +0.010 |
adders | +0.009 |
done | +0.008 |
neys | +0.008 |
measuring | +0.008 |
Neg logit contributions | |
maid | -0.011 |
Qu | -0.010 |
ora | -0.010 |
fish | -0.010 |
bl | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | +8.54e-01 |
Pos logit contributions | |
supposed | +0.025 |
adders | +0.024 |
neys | +0.021 |
online | +0.019 |
measuring | +0.019 |
Neg logit contributions | |
maid | -0.042 |
fish | -0.041 |
Qu | -0.039 |
ora | -0.039 |
bl | -0.038 |
Token | From |
Token ID | 4863 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
supposed | +0.028 |
adders | +0.027 |
By | +0.024 |
about | +0.023 |
online | +0.023 |
Neg logit contributions | |
maid | -0.042 |
fish | -0.039 |
breath | -0.036 |
ora | -0.036 |
bl | -0.036 |
Token | then |
Token ID | 788 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
adders | +0.008 |
supposed | +0.008 |
neys | +0.007 |
By | +0.006 |
measuring | +0.006 |
Neg logit contributions | |
maid | -0.009 |
fish | -0.009 |
ora | -0.009 |
bl | -0.009 |
Qu | -0.009 |
Token | praised |
Token ID | 15342 |
Feature activation | +6.40e-01 |
Pos logit contributions | |
adders | +0.010 |
supposed | +0.009 |
neys | +0.008 |
By | +0.008 |
online | +0.008 |
Neg logit contributions | |
maid | -0.009 |
bl | -0.009 |
fish | -0.009 |
breath | -0.008 |
dolphin | -0.008 |
Token | him |
Token ID | 683 |
Feature activation | +5.19e-02 |
Pos logit contributions | |
supposed | +0.025 |
adders | +0.025 |
neys | +0.022 |
By | +0.021 |
measuring | +0.021 |
Neg logit contributions | |
maid | -0.027 |
fish | -0.025 |
breath | -0.024 |
bl | -0.023 |
ora | -0.023 |
Token | for |
Token ID | 329 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
adders | +0.002 |
supposed | +0.002 |
neys | +0.002 |
emergencies | +0.002 |
measuring | +0.002 |
Neg logit contributions | |
maid | -0.002 |
fish | -0.002 |
breath | -0.002 |
bl | -0.002 |
ora | -0.002 |
Token | his |
Token ID | 465 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
adders | +0.082 |
supposed | +0.078 |
neys | +0.072 |
By | +0.066 |
online | +0.064 |
Neg logit contributions | |
maid | -0.124 |
fish | -0.117 |
bl | -0.114 |
breath | -0.113 |
Qu | -0.112 |
Token | master |
Token ID | 4958 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
maid | +0.038 |
bl | +0.036 |
breath | +0.036 |
fish | +0.035 |
lizard | +0.033 |
Neg logit contributions | |
adders | -0.043 |
supposed | -0.040 |
neys | -0.037 |
By | -0.036 |
Wil | -0.035 |
Token | pieces |
Token ID | 34154 |
Feature activation | +8.09e+00 |
Pos logit contributions | |
maid | +0.073 |
bl | +0.070 |
fish | +0.069 |
breath | +0.068 |
lizard | +0.064 |
Neg logit contributions | |
adders | -0.089 |
supposed | -0.084 |
neys | -0.079 |
Wil | -0.075 |
emergencies | -0.074 |
Token | . |
Token ID | 13 |
Feature activation | +4.20e-02 |
Pos logit contributions | |
adders | +0.339 |
supposed | +0.335 |
emergencies | +0.308 |
Wil | +0.308 |
eps | +0.307 |
Neg logit contributions | |
maid | -0.257 |
fish | -0.255 |
bl | -0.254 |
lizard | -0.231 |
breath | -0.225 |
Token |
|
Token ID | 198 |
Feature activation | +5.01e-01 |
Pos logit contributions | |
supposed | +0.002 |
adders | +0.002 |
neys | +0.001 |
eps | +0.001 |
emergencies | +0.001 |
Neg logit contributions | |
maid | -0.002 |
ora | -0.002 |
fish | -0.002 |
Qu | -0.002 |
bl | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +5.58e-01 |
Pos logit contributions | |
adders | +0.016 |
supposed | +0.016 |
neys | +0.013 |
online | +0.012 |
emergencies | +0.012 |
Neg logit contributions | |
maid | -0.024 |
fish | -0.023 |
Qu | -0.023 |
ora | -0.022 |
bl | -0.022 |
Token | Jack |
Token ID | 14295 |
Feature activation | -4.03e-01 |
Pos logit contributions | |
supposed | +0.022 |
adders | +0.021 |
neys | +0.018 |
about | +0.018 |
By | +0.018 |
Neg logit contributions | |
maid | -0.024 |
fish | -0.022 |
ora | -0.021 |
Qu | -0.021 |
bl | -0.021 |
Token | stopped |
Token ID | 5025 |
Feature activation | +3.32e+00 |
Pos logit contributions | |
maid | +0.017 |
fish | +0.016 |
Qu | +0.016 |
ora | +0.015 |
bl | +0.015 |
Neg logit contributions | |
supposed | -0.015 |
adders | -0.015 |
neys | -0.013 |
about | -0.013 |
online | -0.013 |
Token | boys |
Token ID | 6510 |
Feature activation | +3.38e+00 |
Pos logit contributions | |
adders | +0.044 |
supposed | +0.042 |
neys | +0.038 |
By | +0.036 |
online | +0.034 |
Neg logit contributions | |
maid | -0.044 |
bl | -0.043 |
dolphin | -0.041 |
breath | -0.041 |
fish | -0.041 |
Token | picked |
Token ID | 6497 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
supposed | +0.129 |
adders | +0.128 |
neys | +0.111 |
By | +0.107 |
online | +0.107 |
Neg logit contributions | |
maid | -0.143 |
fish | -0.137 |
Qu | -0.128 |
bl | -0.128 |
ora | -0.127 |
Token | up |
Token ID | 510 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
maid | +0.087 |
fish | +0.082 |
bl | +0.079 |
ora | +0.078 |
dolphin | +0.078 |
Neg logit contributions | |
adders | -0.059 |
supposed | -0.059 |
neys | -0.052 |
By | -0.048 |
emergencies | -0.047 |
Token | their |
Token ID | 511 |
Feature activation | +6.50e-02 |
Pos logit contributions | |
supposed | +0.066 |
adders | +0.064 |
neys | +0.055 |
By | +0.053 |
emergencies | +0.051 |
Neg logit contributions | |
maid | -0.083 |
fish | -0.080 |
bl | -0.076 |
ora | -0.074 |
dolphin | -0.074 |
Token | respective |
Token ID | 11756 |
Feature activation | +2.73e+00 |
Pos logit contributions | |
adders | +0.003 |
supposed | +0.003 |
By | +0.002 |
neys | +0.002 |
iers | +0.002 |
Neg logit contributions | |
maid | -0.003 |
bl | -0.002 |
breath | -0.002 |
fish | -0.002 |
lizard | -0.002 |
Token | tele |
Token ID | 5735 |
Feature activation | +8.06e+00 |
Pos logit contributions | |
adders | +0.099 |
supposed | +0.099 |
neys | +0.083 |
By | +0.082 |
online | +0.081 |
Neg logit contributions | |
maid | -0.123 |
fish | -0.113 |
bl | -0.110 |
ora | -0.109 |
breath | -0.109 |
Token | phones |
Token ID | 9708 |
Feature activation | +3.47e+00 |
Pos logit contributions | |
supposed | +0.291 |
adders | +0.283 |
about | +0.240 |
By | +0.238 |
done | +0.236 |
Neg logit contributions | |
maid | -0.370 |
Qu | -0.333 |
fish | -0.330 |
bl | -0.327 |
ora | -0.325 |
Token | and |
Token ID | 290 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
adders | +0.138 |
supposed | +0.137 |
neys | +0.121 |
emergencies | +0.119 |
By | +0.112 |
Neg logit contributions | |
maid | -0.146 |
fish | -0.142 |
bl | -0.133 |
breath | -0.127 |
ora | -0.125 |
Token | said |
Token ID | 531 |
Feature activation | +2.44e+00 |
Pos logit contributions | |
adders | +0.010 |
supposed | +0.010 |
neys | +0.009 |
By | +0.008 |
online | +0.008 |
Neg logit contributions | |
maid | -0.010 |
fish | -0.009 |
bl | -0.009 |
Qu | -0.009 |
ora | -0.009 |
Token | " |
Token ID | 366 |
Feature activation | -8.26e-01 |
Pos logit contributions | |
supposed | +0.105 |
adders | +0.101 |
neys | +0.093 |
emergencies | +0.086 |
online | +0.085 |
Neg logit contributions | |
maid | -0.096 |
fish | -0.090 |
bl | -0.085 |
Qu | -0.084 |
ora | -0.082 |
Token | good |
Token ID | 11274 |
Feature activation | +4.82e-01 |
Pos logit contributions | |
maid | +0.030 |
fish | +0.029 |
Qu | +0.028 |
ora | +0.027 |
bl | +0.026 |
Neg logit contributions | |
adders | -0.036 |
supposed | -0.036 |
neys | -0.031 |
about | -0.031 |
online | -0.031 |
Token | tools |
Token ID | 4899 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
supposed | +0.014 |
adders | +0.014 |
neys | +0.012 |
online | +0.012 |
By | +0.012 |
Neg logit contributions | |
maid | -0.011 |
fish | -0.010 |
ora | -0.009 |
breath | -0.009 |
bl | -0.009 |
Token | to |
Token ID | 284 |
Feature activation | -4.97e-01 |
Pos logit contributions | |
adders | +0.072 |
supposed | +0.068 |
emergencies | +0.064 |
neys | +0.062 |
elev | +0.062 |
Neg logit contributions | |
fish | -0.062 |
maid | -0.057 |
bl | -0.055 |
breath | -0.050 |
dolphin | -0.049 |
Token | create |
Token ID | 2251 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
maid | +0.019 |
fish | +0.018 |
Qu | +0.017 |
ora | +0.017 |
bl | +0.017 |
Neg logit contributions | |
supposed | -0.021 |
adders | -0.021 |
neys | -0.018 |
By | -0.018 |
about | -0.018 |
Token | his |
Token ID | 465 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
maid | +0.077 |
fish | +0.073 |
bl | +0.068 |
breath | +0.066 |
ora | +0.064 |
Neg logit contributions | |
supposed | -0.073 |
adders | -0.073 |
By | -0.062 |
neys | -0.061 |
Wil | -0.061 |
Token | master |
Token ID | 4958 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
maid | +0.041 |
fish | +0.040 |
breath | +0.039 |
bl | +0.037 |
lizard | +0.037 |
Neg logit contributions | |
adders | -0.054 |
supposed | -0.052 |
Wil | -0.045 |
chairs | -0.045 |
By | -0.045 |
Token | pieces |
Token ID | 34154 |
Feature activation | +8.02e+00 |
Pos logit contributions | |
fish | +0.078 |
maid | +0.077 |
bl | +0.073 |
breath | +0.070 |
copy | +0.066 |
Neg logit contributions | |
adders | -0.082 |
supposed | -0.077 |
Wil | -0.076 |
By | -0.071 |
neys | -0.070 |
Token | . |
Token ID | 13 |
Feature activation | +3.88e-01 |
Pos logit contributions | |
adders | +0.337 |
supposed | +0.334 |
Wil | +0.321 |
emergencies | +0.301 |
traffic | +0.299 |
Neg logit contributions | |
maid | -0.262 |
fish | -0.258 |
bl | -0.249 |
ora | -0.215 |
mer | -0.214 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.44e+00 |
Pos logit contributions | |
adders | +0.015 |
supposed | +0.015 |
eps | +0.013 |
done | +0.013 |
emergencies | +0.013 |
Neg logit contributions | |
Qu | -0.014 |
maid | -0.013 |
of | -0.013 |
ora | -0.012 |
lizard | -0.012 |
Token | Once |
Token ID | 7454 |
Feature activation | -3.04e+00 |
Pos logit contributions | |
maid | +0.155 |
fish | +0.150 |
Qu | +0.148 |
ora | +0.145 |
bl | +0.141 |
Neg logit contributions | |
adders | -0.116 |
supposed | -0.114 |
neys | -0.096 |
online | -0.094 |
measuring | -0.092 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
maid | +0.107 |
fish | +0.103 |
ora | +0.094 |
breath | +0.091 |
bl | +0.090 |
Neg logit contributions | |
supposed | -0.136 |
adders | -0.136 |
neys | -0.123 |
emergencies | -0.118 |
driving | -0.115 |
Token | a |
Token ID | 257 |
Feature activation | -8.02e-01 |
Pos logit contributions | |
maid | +0.075 |
fish | +0.072 |
Qu | +0.070 |
ora | +0.069 |
bl | +0.069 |
Neg logit contributions | |
adders | -0.032 |
supposed | -0.032 |
neys | -0.026 |
online | -0.024 |
By | -0.024 |
Token | dog |
Token ID | 3290 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
maid | +0.095 |
fish | +0.085 |
Qu | +0.084 |
ora | +0.083 |
breath | +0.078 |
Neg logit contributions | |
adders | -0.100 |
supposed | -0.100 |
measuring | -0.096 |
traffic | -0.095 |
online | -0.093 |
Token | named |
Token ID | 3706 |
Feature activation | +3.19e-01 |
Pos logit contributions | |
maid | +0.119 |
ora | +0.109 |
fish | +0.108 |
breath | +0.106 |
Qu | +0.105 |
Neg logit contributions | |
supposed | -0.056 |
adders | -0.053 |
about | -0.051 |
driving | -0.047 |
neys | -0.046 |
Token | Sleep |
Token ID | 17376 |
Feature activation | -6.51e+00 |
Pos logit contributions | |
supposed | +0.011 |
adders | +0.011 |
neys | +0.009 |
online | +0.008 |
done | +0.008 |
Neg logit contributions | |
maid | -0.015 |
Qu | -0.014 |
fish | -0.014 |
bl | -0.014 |
dolphin | -0.013 |
Token | y |
Token ID | 88 |
Feature activation | -3.33e+00 |
Pos logit contributions | |
maid | +0.283 |
jugg | +0.259 |
breath | +0.257 |
fish | +0.255 |
snail | +0.250 |
Neg logit contributions | |
supposed | -0.210 |
late | -0.197 |
about | -0.196 |
outside | -0.194 |
neys | -0.172 |
Token | . |
Token ID | 13 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
maid | +0.216 |
breath | +0.209 |
ora | +0.208 |
Qu | +0.194 |
fish | +0.194 |
Neg logit contributions | |
about | -0.034 |
driving | -0.029 |
supposed | -0.023 |
outside | -0.021 |
much | -0.018 |
Token | Sleep |
Token ID | 17376 |
Feature activation | -6.56e+00 |
Pos logit contributions | |
maid | +0.008 |
fish | +0.007 |
ora | +0.007 |
breath | +0.006 |
ho | +0.006 |
Neg logit contributions | |
There | -0.005 |
By | -0.005 |
about | -0.005 |
late | -0.005 |
supposed | -0.005 |
Token | y |
Token ID | 88 |
Feature activation | -3.33e+00 |
Pos logit contributions | |
maid | +0.299 |
jugg | +0.291 |
snail | +0.283 |
breath | +0.281 |
groom | +0.278 |
Neg logit contributions | |
about | -0.202 |
supposed | -0.193 |
late | -0.183 |
outside | -0.177 |
done | -0.172 |
Token | was |
Token ID | 373 |
Feature activation | +5.13e-01 |
Pos logit contributions | |
maid | +0.183 |
breath | +0.172 |
ora | +0.163 |
fish | +0.159 |
Qu | +0.157 |
Neg logit contributions | |
about | -0.079 |
supposed | -0.077 |
done | -0.068 |
adders | -0.067 |
driving | -0.064 |
Token | a |
Token ID | 257 |
Feature activation | -9.03e-01 |
Pos logit contributions | |
supposed | +0.012 |
adders | +0.010 |
about | +0.009 |
driving | +0.008 |
done | +0.008 |
Neg logit contributions | |
maid | -0.031 |
fish | -0.030 |
ora | -0.029 |
Qu | -0.028 |
breath | -0.027 |
Token | very |
Token ID | 845 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
maid | +0.052 |
ora | +0.049 |
fish | +0.047 |
breath | +0.046 |
Qu | +0.045 |
Neg logit contributions | |
supposed | -0.021 |
measuring | -0.017 |
late | -0.016 |
driving | -0.016 |
ready | -0.015 |
Token | sleepy |
Token ID | 42023 |
Feature activation | -7.11e-01 |
Pos logit contributions | |
supposed | +0.007 |
adders | +0.006 |
about | +0.006 |
done | +0.006 |
online | +0.005 |
Neg logit contributions | |
maid | -0.016 |
fish | -0.014 |
Qu | -0.014 |
ora | -0.014 |
breath | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | +3.32e-01 |
Pos logit contributions | |
maid | +0.006 |
fish | +0.005 |
ora | +0.005 |
Qu | +0.005 |
breath | +0.005 |
Neg logit contributions | |
supposed | -0.002 |
adders | -0.002 |
about | -0.002 |
By | -0.002 |
neys | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +3.84e-01 |
Pos logit contributions | |
supposed | +0.007 |
adders | +0.007 |
about | +0.006 |
By | +0.005 |
neys | +0.005 |
Neg logit contributions | |
maid | -0.020 |
fish | -0.019 |
Qu | -0.018 |
ora | -0.018 |
bl | -0.018 |
Token | One |
Token ID | 3198 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
By | +0.006 |
about | +0.006 |
supposed | +0.006 |
late | +0.006 |
traffic | +0.005 |
Neg logit contributions | |
maid | -0.025 |
breath | -0.022 |
fish | -0.022 |
ora | -0.022 |
snail | -0.020 |
Token | day |
Token ID | 1110 |
Feature activation | -7.09e-01 |
Pos logit contributions | |
supposed | +0.034 |
adders | +0.033 |
neys | +0.026 |
about | +0.025 |
online | +0.025 |
Neg logit contributions | |
maid | -0.076 |
fish | -0.073 |
Qu | -0.070 |
ora | -0.070 |
bl | -0.068 |
Token | , |
Token ID | 11 |
Feature activation | -3.88e-01 |
Pos logit contributions | |
maid | +0.040 |
ora | +0.038 |
fish | +0.038 |
Qu | +0.036 |
breath | +0.035 |
Neg logit contributions | |
supposed | -0.016 |
adders | -0.015 |
about | -0.014 |
By | -0.013 |
measuring | -0.013 |
Token | Sleep |
Token ID | 17376 |
Feature activation | -6.55e+00 |
Pos logit contributions | |
maid | +0.022 |
ora | +0.021 |
fish | +0.020 |
Qu | +0.019 |
breath | +0.019 |
Neg logit contributions | |
supposed | -0.008 |
driving | -0.008 |
about | -0.008 |
adders | -0.008 |
late | -0.008 |
Token | y |
Token ID | 88 |
Feature activation | -3.29e+00 |
Pos logit contributions | |
maid | +0.265 |
breath | +0.239 |
Qu | +0.238 |
fish | +0.238 |
snail | +0.238 |
Neg logit contributions | |
supposed | -0.242 |
late | -0.231 |
about | -0.227 |
done | -0.213 |
outside | -0.208 |
Token | went |
Token ID | 1816 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
maid | +0.139 |
fish | +0.125 |
Qu | +0.123 |
breath | +0.121 |
ora | +0.121 |
Neg logit contributions | |
supposed | -0.124 |
adders | -0.119 |
about | -0.113 |
done | -0.109 |
By | -0.105 |
Token | for |
Token ID | 329 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
supposed | +0.023 |
adders | +0.022 |
about | +0.017 |
online | +0.017 |
By | +0.016 |
Neg logit contributions | |
maid | -0.071 |
fish | -0.067 |
Qu | -0.065 |
ora | -0.065 |
bl | -0.062 |
Token | a |
Token ID | 257 |
Feature activation | -8.88e-01 |
Pos logit contributions | |
supposed | +0.047 |
adders | +0.043 |
about | +0.035 |
driving | +0.033 |
By | +0.032 |
Neg logit contributions | |
maid | -0.147 |
fish | -0.139 |
ora | -0.136 |
Qu | -0.133 |
bl | -0.128 |
Token | walk |
Token ID | 2513 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
maid | +0.044 |
ora | +0.040 |
fish | +0.039 |
Qu | +0.039 |
bl | +0.037 |
Neg logit contributions | |
supposed | -0.028 |
adders | -0.025 |
about | -0.023 |
measuring | -0.023 |
online | -0.023 |
Token | day |
Token ID | 1110 |
Feature activation | -7.35e-01 |
Pos logit contributions | |
maid | +0.069 |
fish | +0.067 |
Qu | +0.064 |
ora | +0.064 |
bl | +0.063 |
Neg logit contributions | |
adders | -0.032 |
supposed | -0.032 |
neys | -0.025 |
online | -0.024 |
By | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
maid | +0.040 |
fish | +0.037 |
Qu | +0.036 |
ora | +0.036 |
bl | +0.035 |
Neg logit contributions | |
supposed | -0.020 |
adders | -0.019 |
about | -0.016 |
online | -0.016 |
By | -0.016 |
Token | all |
Token ID | 477 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
supposed | +0.004 |
about | +0.003 |
done | +0.003 |
adders | +0.003 |
going | +0.003 |
Neg logit contributions | |
maid | -0.005 |
ora | -0.004 |
fish | -0.004 |
Qu | -0.004 |
snail | -0.004 |
Token | night |
Token ID | 1755 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
maid | +0.071 |
fish | +0.066 |
ora | +0.064 |
Qu | +0.064 |
bl | +0.062 |
Neg logit contributions | |
supposed | -0.029 |
adders | -0.027 |
about | -0.023 |
done | -0.022 |
neys | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -9.92e-02 |
Pos logit contributions | |
adders | +0.000 |
supposed | +0.000 |
neys | +0.000 |
online | +0.000 |
measuring | +0.000 |
Neg logit contributions | |
maid | -0.001 |
fish | -0.001 |
Qu | -0.001 |
ora | -0.001 |
bl | -0.001 |
Token | Sleep |
Token ID | 17376 |
Feature activation | -6.53e+00 |
Pos logit contributions | |
maid | +0.006 |
fish | +0.006 |
ora | +0.005 |
Qu | +0.005 |
breath | +0.005 |
Neg logit contributions | |
supposed | -0.002 |
adders | -0.002 |
about | -0.002 |
By | -0.002 |
neys | -0.002 |
Token | y |
Token ID | 88 |
Feature activation | -3.27e+00 |
Pos logit contributions | |
maid | +0.291 |
snail | +0.277 |
jugg | +0.270 |
breath | +0.269 |
Qu | +0.267 |
Neg logit contributions | |
about | -0.212 |
supposed | -0.209 |
late | -0.198 |
done | -0.193 |
outside | -0.180 |
Token | had |
Token ID | 550 |
Feature activation | +9.63e-01 |
Pos logit contributions | |
maid | +0.156 |
ora | +0.140 |
Qu | +0.139 |
fish | +0.137 |
breath | +0.136 |
Neg logit contributions | |
supposed | -0.106 |
adders | -0.101 |
about | -0.099 |
done | -0.096 |
By | -0.089 |
Token | a |
Token ID | 257 |
Feature activation | -8.46e-01 |
Pos logit contributions | |
supposed | +0.023 |
adders | +0.022 |
neys | +0.018 |
about | +0.017 |
online | +0.017 |
Neg logit contributions | |
maid | -0.055 |
fish | -0.053 |
Qu | -0.051 |
ora | -0.051 |
bl | -0.050 |
Token | big |
Token ID | 1263 |
Feature activation | +4.92e-02 |
Pos logit contributions | |
maid | +0.048 |
ora | +0.046 |
fish | +0.046 |
Qu | +0.044 |
of | +0.042 |
Neg logit contributions | |
supposed | -0.020 |
measuring | -0.016 |
done | -0.015 |
about | -0.015 |
best | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | -3.97e-01 |
Pos logit contributions | |
supposed | +0.002 |
adders | +0.001 |
about | +0.001 |
done | +0.001 |
measuring | +0.001 |
Neg logit contributions | |
maid | -0.002 |
fish | -0.002 |
ora | -0.002 |
Qu | -0.002 |
snail | -0.002 |
Token | night |
Token ID | 1755 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
maid | +0.049 |
fish | +0.046 |
ora | +0.044 |
Qu | +0.044 |
bl | +0.044 |
Neg logit contributions | |
adders | -0.041 |
supposed | -0.040 |
neys | -0.035 |
online | -0.034 |
By | -0.033 |
Token | , |
Token ID | 11 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
adders | +0.011 |
supposed | +0.010 |
neys | +0.010 |
online | +0.009 |
eps | +0.009 |
Neg logit contributions | |
maid | -0.010 |
fish | -0.009 |
breath | -0.009 |
bl | -0.009 |
dolphin | -0.009 |
Token | sweet |
Token ID | 6029 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
maid | +0.009 |
fish | +0.008 |
bl | +0.008 |
Qu | +0.008 |
ora | +0.007 |
Neg logit contributions | |
adders | -0.008 |
supposed | -0.008 |
neys | -0.008 |
online | -0.007 |
By | -0.007 |
Token | ie |
Token ID | 494 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
maid | +0.037 |
fish | +0.034 |
Qu | +0.032 |
ora | +0.029 |
bl | +0.028 |
Neg logit contributions | |
supposed | -0.092 |
adders | -0.090 |
about | -0.084 |
neys | -0.082 |
done | -0.082 |
Token | . |
Token ID | 13 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
maid | +0.078 |
fish | +0.074 |
Qu | +0.070 |
ora | +0.070 |
bl | +0.069 |
Neg logit contributions | |
supposed | -0.059 |
adders | -0.059 |
neys | -0.050 |
online | -0.048 |
By | -0.048 |
Token | Sleep |
Token ID | 17376 |
Feature activation | -6.53e+00 |
Pos logit contributions | |
supposed | +0.007 |
adders | +0.007 |
neys | +0.006 |
online | +0.006 |
done | +0.006 |
Neg logit contributions | |
maid | -0.007 |
fish | -0.007 |
Qu | -0.007 |
ora | -0.007 |
bl | -0.006 |
Token | tight |
Token ID | 5381 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
maid | +0.324 |
fish | +0.309 |
Qu | +0.309 |
snail | +0.304 |
jugg | +0.293 |
Neg logit contributions | |
supposed | -0.182 |
about | -0.177 |
done | -0.164 |
late | -0.154 |
adders | -0.154 |
Token | ." |
Token ID | 526 |
Feature activation | -8.34e-01 |
Pos logit contributions | |
maid | +0.050 |
fish | +0.047 |
Qu | +0.045 |
ora | +0.044 |
bl | +0.044 |
Neg logit contributions | |
supposed | -0.039 |
adders | -0.039 |
neys | -0.033 |
online | -0.032 |
By | -0.032 |
Token | He |
Token ID | 679 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
maid | +0.041 |
fish | +0.038 |
Qu | +0.037 |
ora | +0.037 |
bl | +0.037 |
Neg logit contributions | |
supposed | -0.028 |
adders | -0.027 |
neys | -0.023 |
measuring | -0.022 |
online | -0.022 |
Token | tucked |
Token ID | 29779 |
Feature activation | -9.43e-01 |
Pos logit contributions | |
maid | +0.049 |
ora | +0.045 |
fish | +0.044 |
Qu | +0.043 |
snail | +0.043 |
Neg logit contributions | |
supposed | -0.034 |
about | -0.034 |
adders | -0.033 |
done | -0.032 |
counted | -0.030 |
Token | her |
Token ID | 607 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
maid | +0.044 |
fish | +0.042 |
Qu | +0.040 |
ora | +0.039 |
bl | +0.038 |
Neg logit contributions | |
adders | -0.031 |
supposed | -0.031 |
about | -0.027 |
online | -0.027 |
neys | -0.026 |
Token | was |
Token ID | 373 |
Feature activation | +5.30e-01 |
Pos logit contributions | |
maid | +0.012 |
fish | +0.011 |
ora | +0.010 |
Qu | +0.010 |
breath | +0.010 |
Neg logit contributions | |
supposed | -0.008 |
adders | -0.008 |
about | -0.007 |
done | -0.007 |
measuring | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | -9.20e-01 |
Pos logit contributions | |
adders | +0.018 |
neys | +0.018 |
supposed | +0.018 |
cases | +0.016 |
chairs | +0.016 |
Neg logit contributions | |
Qu | -0.022 |
maid | -0.021 |
dolphin | -0.021 |
bl | -0.021 |
fish | -0.021 |
Token | little |
Token ID | 1310 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
maid | +0.041 |
fish | +0.038 |
ora | +0.037 |
Qu | +0.036 |
breath | +0.036 |
Neg logit contributions | |
supposed | -0.031 |
adders | -0.029 |
measuring | -0.028 |
driving | -0.027 |
online | -0.027 |
Token | dog |
Token ID | 3290 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
maid | +0.095 |
fish | +0.085 |
Qu | +0.084 |
ora | +0.083 |
breath | +0.078 |
Neg logit contributions | |
adders | -0.100 |
supposed | -0.100 |
measuring | -0.096 |
traffic | -0.095 |
online | -0.093 |
Token | named |
Token ID | 3706 |
Feature activation | +3.19e-01 |
Pos logit contributions | |
maid | +0.119 |
ora | +0.109 |
fish | +0.108 |
breath | +0.106 |
Qu | +0.105 |
Neg logit contributions | |
supposed | -0.056 |
adders | -0.053 |
about | -0.051 |
driving | -0.047 |
neys | -0.046 |
Token | Sleep |
Token ID | 17376 |
Feature activation | -6.51e+00 |
Pos logit contributions | |
supposed | +0.011 |
adders | +0.011 |
neys | +0.009 |
online | +0.008 |
done | +0.008 |
Neg logit contributions | |
maid | -0.015 |
Qu | -0.014 |
fish | -0.014 |
bl | -0.014 |
dolphin | -0.013 |
Token | y |
Token ID | 88 |
Feature activation | -3.33e+00 |
Pos logit contributions | |
maid | +0.283 |
jugg | +0.259 |
breath | +0.257 |
fish | +0.255 |
snail | +0.250 |
Neg logit contributions | |
supposed | -0.210 |
late | -0.197 |
about | -0.196 |
outside | -0.194 |
neys | -0.172 |
Token | . |
Token ID | 13 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
maid | +0.216 |
breath | +0.209 |
ora | +0.208 |
Qu | +0.194 |
fish | +0.194 |
Neg logit contributions | |
about | -0.034 |
driving | -0.029 |
supposed | -0.023 |
outside | -0.021 |
much | -0.018 |
Token | Sleep |
Token ID | 17376 |
Feature activation | -6.56e+00 |
Pos logit contributions | |
maid | +0.008 |
fish | +0.007 |
ora | +0.007 |
breath | +0.006 |
ho | +0.006 |
Neg logit contributions | |
There | -0.005 |
By | -0.005 |
about | -0.005 |
late | -0.005 |
supposed | -0.005 |
Token | y |
Token ID | 88 |
Feature activation | -3.33e+00 |
Pos logit contributions | |
maid | +0.299 |
jugg | +0.291 |
snail | +0.283 |
breath | +0.281 |
groom | +0.278 |
Neg logit contributions | |
about | -0.202 |
supposed | -0.193 |
late | -0.183 |
outside | -0.177 |
done | -0.172 |
Token | was |
Token ID | 373 |
Feature activation | +5.13e-01 |
Pos logit contributions | |
maid | +0.183 |
breath | +0.172 |
ora | +0.163 |
fish | +0.159 |
Qu | +0.157 |
Neg logit contributions | |
about | -0.079 |
supposed | -0.077 |
done | -0.068 |
adders | -0.067 |
driving | -0.064 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
supposed | +0.006 |
adders | +0.006 |
about | +0.005 |
done | +0.005 |
neys | +0.005 |
Neg logit contributions | |
maid | -0.008 |
fish | -0.007 |
ora | -0.007 |
Qu | -0.007 |
breath | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | -7.73e-01 |
Pos logit contributions | |
supposed | +0.031 |
adders | +0.030 |
about | +0.024 |
neys | +0.023 |
online | +0.022 |
Neg logit contributions | |
maid | -0.087 |
fish | -0.083 |
ora | -0.080 |
Qu | -0.080 |
breath | -0.078 |
Token | big |
Token ID | 1263 |
Feature activation | +7.71e-02 |
Pos logit contributions | |
maid | +0.038 |
ora | +0.035 |
fish | +0.035 |
Qu | +0.034 |
breath | +0.033 |
Neg logit contributions | |
supposed | -0.023 |
measuring | -0.020 |
adders | -0.020 |
traffic | -0.019 |
driving | -0.019 |
Token | ball |
Token ID | 2613 |
Feature activation | -6.01e-01 |
Pos logit contributions | |
supposed | +0.003 |
adders | +0.002 |
about | +0.002 |
measuring | +0.002 |
traffic | +0.002 |
Neg logit contributions | |
maid | -0.003 |
fish | -0.003 |
ora | -0.003 |
Qu | -0.003 |
breath | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +9.63e-02 |
Pos logit contributions | |
maid | +0.034 |
fish | +0.031 |
ora | +0.031 |
Qu | +0.031 |
breath | +0.030 |
Neg logit contributions | |
supposed | -0.015 |
adders | -0.015 |
about | -0.012 |
driving | -0.012 |
online | -0.012 |
Token | Sleep |
Token ID | 17376 |
Feature activation | -6.50e+00 |
Pos logit contributions | |
adders | +0.002 |
supposed | +0.002 |
about | +0.002 |
By | +0.002 |
There | +0.002 |
Neg logit contributions | |
maid | -0.006 |
fish | -0.006 |
ora | -0.005 |
Qu | -0.005 |
breath | -0.005 |
Token | y |
Token ID | 88 |
Feature activation | -3.20e+00 |
Pos logit contributions | |
maid | +0.246 |
snail | +0.237 |
Qu | +0.235 |
jugg | +0.227 |
groom | +0.222 |
Neg logit contributions | |
supposed | -0.246 |
about | -0.241 |
late | -0.229 |
done | -0.225 |
outside | -0.218 |
Token | loved |
Token ID | 6151 |
Feature activation | -8.18e-01 |
Pos logit contributions | |
maid | +0.145 |
ora | +0.131 |
Qu | +0.127 |
fish | +0.126 |
breath | +0.122 |
Neg logit contributions | |
supposed | -0.104 |
about | -0.103 |
adders | -0.098 |
driving | -0.095 |
done | -0.095 |
Token | balls |
Token ID | 11333 |
Feature activation | +3.41e+00 |
Pos logit contributions | |
maid | +0.053 |
ora | +0.049 |
fish | +0.048 |
Qu | +0.046 |
breath | +0.044 |
Neg logit contributions | |
supposed | -0.017 |
about | -0.014 |
adders | -0.013 |
driving | -0.013 |
going | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
adders | +0.123 |
supposed | +0.120 |
neys | +0.109 |
emergencies | +0.106 |
By | +0.103 |
Neg logit contributions | |
fish | -0.147 |
maid | -0.145 |
bl | -0.137 |
Qu | -0.130 |
breath | -0.130 |
Token | He |
Token ID | 679 |
Feature activation | -3.46e-01 |
Pos logit contributions | |
adders | +0.003 |
supposed | +0.003 |
about | +0.003 |
By | +0.003 |
There | +0.003 |
Neg logit contributions | |
maid | -0.008 |
fish | -0.008 |
ora | -0.007 |
Qu | -0.007 |
breath | -0.007 |
Token | " |
Token ID | 366 |
Feature activation | -8.85e-01 |
Pos logit contributions | |
maid | +0.008 |
fish | +0.008 |
bl | +0.008 |
Qu | +0.008 |
dolphin | +0.008 |
Neg logit contributions | |
supposed | -0.008 |
adders | -0.007 |
neys | -0.007 |
done | -0.006 |
emergencies | -0.006 |
Token | It |
Token ID | 1026 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
maid | +0.037 |
fish | +0.033 |
breath | +0.033 |
bl | +0.031 |
ora | +0.030 |
Neg logit contributions | |
supposed | -0.033 |
adders | -0.032 |
By | -0.030 |
neys | -0.030 |
online | -0.029 |
Token | 's |
Token ID | 338 |
Feature activation | -8.08e-01 |
Pos logit contributions | |
maid | +0.070 |
fish | +0.068 |
bl | +0.065 |
ora | +0.063 |
Qu | +0.062 |
Neg logit contributions | |
adders | -0.053 |
supposed | -0.051 |
neys | -0.044 |
online | -0.042 |
emergencies | -0.042 |
Token | easy |
Token ID | 2562 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
maid | +0.033 |
bl | +0.032 |
breath | +0.030 |
fish | +0.030 |
dolphin | +0.029 |
Neg logit contributions | |
adders | -0.032 |
neys | -0.027 |
cases | -0.025 |
supposed | -0.025 |
chairs | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +3.25e-01 |
Pos logit contributions | |
adders | +0.044 |
supposed | +0.041 |
neys | +0.038 |
eps | +0.036 |
By | +0.035 |
Neg logit contributions | |
fish | -0.058 |
maid | -0.058 |
bl | -0.053 |
breath | -0.052 |
ora | -0.052 |
Token | Sleep |
Token ID | 17376 |
Feature activation | -6.49e+00 |
Pos logit contributions | |
supposed | +0.014 |
adders | +0.013 |
done | +0.012 |
neys | +0.011 |
online | +0.011 |
Neg logit contributions | |
maid | -0.013 |
Qu | -0.012 |
fish | -0.012 |
ora | -0.012 |
bl | -0.011 |
Token | is |
Token ID | 318 |
Feature activation | +4.46e-02 |
Pos logit contributions | |
maid | +0.307 |
Qu | +0.292 |
fish | +0.284 |
snail | +0.274 |
bl | +0.273 |
Neg logit contributions | |
supposed | -0.204 |
about | -0.189 |
adders | -0.182 |
done | -0.173 |
neys | -0.169 |
Token | when |
Token ID | 618 |
Feature activation | +6.56e-01 |
Pos logit contributions | |
adders | +0.002 |
neys | +0.001 |
supposed | +0.001 |
chairs | +0.001 |
cases | +0.001 |
Neg logit contributions | |
maid | -0.002 |
fish | -0.002 |
bl | -0.002 |
breath | -0.002 |
dolphin | -0.002 |
Token | you |
Token ID | 345 |
Feature activation | +9.10e-01 |
Pos logit contributions | |
supposed | +0.027 |
adders | +0.027 |
neys | +0.024 |
online | +0.023 |
about | +0.023 |
Neg logit contributions | |
maid | -0.025 |
fish | -0.024 |
Qu | -0.022 |
ora | -0.022 |
bl | -0.022 |
Token | close |
Token ID | 1969 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
supposed | +0.036 |
adders | +0.036 |
about | +0.031 |
done | +0.031 |
neys | +0.031 |
Neg logit contributions | |
maid | -0.037 |
fish | -0.034 |
Qu | -0.033 |
ora | -0.032 |
bl | -0.031 |
Token | your |
Token ID | 534 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
maid | +0.063 |
fish | +0.060 |
ora | +0.056 |
Qu | +0.056 |
bl | +0.055 |
Neg logit contributions | |
supposed | -0.053 |
adders | -0.053 |
neys | -0.046 |
online | -0.044 |
By | -0.044 |
Token | L |
Token ID | 43 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
supposed | +0.019 |
adders | +0.018 |
neys | +0.016 |
about | +0.016 |
online | +0.016 |
Neg logit contributions | |
maid | -0.015 |
fish | -0.014 |
Qu | -0.014 |
ora | -0.014 |
bl | -0.013 |
Token | ily |
Token ID | 813 |
Feature activation | -9.75e-01 |
Pos logit contributions | |
maid | +0.064 |
fish | +0.061 |
Qu | +0.059 |
breath | +0.058 |
ora | +0.057 |
Neg logit contributions | |
supposed | -0.036 |
adders | -0.036 |
about | -0.029 |
neys | -0.029 |
online | -0.028 |
Token | took |
Token ID | 1718 |
Feature activation | -8.68e-01 |
Pos logit contributions | |
maid | +0.038 |
fish | +0.036 |
Qu | +0.034 |
ora | +0.033 |
bl | +0.032 |
Neg logit contributions | |
supposed | -0.040 |
adders | -0.040 |
neys | -0.035 |
about | -0.035 |
online | -0.034 |
Token | a |
Token ID | 257 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
breath | +0.035 |
bl | +0.034 |
maid | +0.034 |
ora | +0.032 |
Qu | +0.031 |
Neg logit contributions | |
supposed | -0.034 |
adders | -0.032 |
neys | -0.031 |
By | -0.027 |
driving | -0.027 |
Token | deep |
Token ID | 2769 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
maid | +0.049 |
fish | +0.046 |
Qu | +0.043 |
ora | +0.043 |
bl | +0.041 |
Neg logit contributions | |
supposed | -0.063 |
adders | -0.062 |
neys | -0.055 |
about | -0.054 |
online | -0.054 |
Token | breath |
Token ID | 8033 |
Feature activation | -6.48e+00 |
Pos logit contributions | |
maid | +0.024 |
Qu | +0.020 |
fish | +0.019 |
ora | +0.019 |
wrists | +0.016 |
Neg logit contributions | |
measuring | -0.054 |
late | -0.052 |
done | -0.052 |
adders | -0.051 |
supposed | -0.050 |
Token | and |
Token ID | 290 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
maid | +0.293 |
fish | +0.280 |
Qu | +0.266 |
ora | +0.264 |
bl | +0.262 |
Neg logit contributions | |
supposed | -0.228 |
adders | -0.225 |
neys | -0.197 |
online | -0.184 |
By | -0.183 |
Token | reached |
Token ID | 4251 |
Feature activation | +6.05e-01 |
Pos logit contributions | |
supposed | +0.009 |
adders | +0.009 |
neys | +0.008 |
By | +0.007 |
online | +0.007 |
Neg logit contributions | |
maid | -0.008 |
fish | -0.008 |
Qu | -0.007 |
bl | -0.007 |
ora | -0.007 |
Token | for |
Token ID | 329 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
supposed | +0.021 |
adders | +0.020 |
neys | +0.020 |
emergencies | +0.018 |
clock | +0.017 |
Neg logit contributions | |
breath | -0.027 |
maid | -0.026 |
fish | -0.025 |
bl | -0.025 |
ora | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
supposed | +0.075 |
adders | +0.074 |
neys | +0.063 |
about | +0.060 |
online | +0.060 |
Neg logit contributions | |
maid | -0.115 |
fish | -0.109 |
Qu | -0.105 |
ora | -0.104 |
breath | -0.104 |
Token | pot |
Token ID | 1787 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
adders | +0.011 |
supposed | +0.011 |
neys | +0.010 |
By | +0.010 |
about | +0.010 |
Neg logit contributions | |
maid | -0.008 |
fish | -0.007 |
bl | -0.007 |
dolphin | -0.007 |
breath | -0.007 |
Token | you |
Token ID | 345 |
Feature activation | +6.25e-01 |
Pos logit contributions | |
maid | +0.035 |
fish | +0.033 |
Qu | +0.032 |
ora | +0.031 |
bl | +0.031 |
Neg logit contributions | |
adders | -0.036 |
supposed | -0.036 |
neys | -0.031 |
online | -0.030 |
about | -0.030 |
Token | too |
Token ID | 1165 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
supposed | +0.022 |
adders | +0.022 |
neys | +0.019 |
online | +0.018 |
By | +0.018 |
Neg logit contributions | |
maid | -0.028 |
fish | -0.027 |
Qu | -0.026 |
ora | -0.025 |
bl | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | -5.86e-01 |
Pos logit contributions | |
maid | +0.005 |
fish | +0.004 |
bl | +0.004 |
breath | +0.004 |
ora | +0.004 |
Neg logit contributions | |
adders | -0.004 |
supposed | -0.004 |
neys | -0.003 |
online | -0.003 |
emergencies | -0.003 |
Token | kids |
Token ID | 3988 |
Feature activation | +3.16e+00 |
Pos logit contributions | |
maid | +0.029 |
fish | +0.028 |
ora | +0.027 |
Qu | +0.026 |
bl | +0.025 |
Neg logit contributions | |
supposed | -0.017 |
adders | -0.017 |
about | -0.015 |
online | -0.014 |
By | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -4.82e-01 |
Pos logit contributions | |
supposed | +0.107 |
adders | +0.106 |
neys | +0.089 |
By | +0.086 |
online | +0.085 |
Neg logit contributions | |
maid | -0.149 |
fish | -0.142 |
Qu | -0.134 |
ora | -0.134 |
bl | -0.134 |
Token | Sleep |
Token ID | 17376 |
Feature activation | -6.43e+00 |
Pos logit contributions | |
maid | +0.021 |
Qu | +0.020 |
fish | +0.020 |
ora | +0.019 |
bl | +0.019 |
Neg logit contributions | |
supposed | -0.017 |
adders | -0.017 |
neys | -0.014 |
done | -0.014 |
online | -0.014 |
Token | well |
Token ID | 880 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
maid | +0.326 |
Qu | +0.306 |
jugg | +0.304 |
fish | +0.303 |
snail | +0.297 |
Neg logit contributions | |
about | -0.171 |
supposed | -0.169 |
done | -0.155 |
late | -0.147 |
neys | -0.140 |
Token | . |
Token ID | 13 |
Feature activation | -4.35e-01 |
Pos logit contributions | |
maid | +0.117 |
fish | +0.114 |
ora | +0.106 |
Qu | +0.105 |
bl | +0.105 |
Neg logit contributions | |
supposed | -0.084 |
adders | -0.083 |
neys | -0.070 |
By | -0.068 |
measuring | -0.067 |
Token | Have |
Token ID | 8192 |
Feature activation | +2.57e-01 |
Pos logit contributions | |
maid | +0.022 |
fish | +0.021 |
Qu | +0.020 |
ora | +0.020 |
bl | +0.020 |
Neg logit contributions | |
supposed | -0.013 |
adders | -0.013 |
neys | -0.011 |
about | -0.010 |
online | -0.010 |
Token | sweet |
Token ID | 6029 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
supposed | +0.004 |
adders | +0.003 |
about | +0.002 |
done | +0.002 |
By | +0.002 |
Neg logit contributions | |
maid | -0.018 |
fish | -0.017 |
ora | -0.016 |
Qu | -0.016 |
breath | -0.016 |
Token | dreams |
Token ID | 10625 |
Feature activation | +8.36e-01 |
Pos logit contributions | |
maid | +0.075 |
fish | +0.070 |
Qu | +0.068 |
ora | +0.066 |
bl | +0.063 |
Neg logit contributions | |
supposed | -0.050 |
adders | -0.048 |
about | -0.043 |
done | -0.041 |
neys | -0.041 |
Token | Good |
Token ID | 10248 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
maid | +0.058 |
fish | +0.053 |
ora | +0.050 |
breath | +0.049 |
bl | +0.049 |
Neg logit contributions | |
supposed | -0.054 |
adders | -0.053 |
neys | -0.047 |
By | -0.046 |
measuring | -0.046 |
Token | night |
Token ID | 1755 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
maid | +0.050 |
fish | +0.048 |
Qu | +0.046 |
ora | +0.045 |
bl | +0.044 |
Neg logit contributions | |
supposed | -0.035 |
adders | -0.035 |
neys | -0.029 |
about | -0.029 |
online | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
adders | +0.011 |
supposed | +0.011 |
neys | +0.010 |
online | +0.009 |
eps | +0.009 |
Neg logit contributions | |
maid | -0.010 |
fish | -0.010 |
breath | -0.009 |
dolphin | -0.009 |
Qu | -0.009 |
Token | drum |
Token ID | 13026 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
maid | +0.008 |
fish | +0.008 |
Qu | +0.007 |
dolphin | +0.007 |
bl | +0.007 |
Neg logit contributions | |
supposed | -0.007 |
adders | -0.007 |
neys | -0.006 |
online | -0.006 |
By | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +3.57e-01 |
Pos logit contributions | |
supposed | +0.004 |
adders | +0.004 |
neys | +0.004 |
online | +0.004 |
By | +0.004 |
Neg logit contributions | |
maid | -0.005 |
fish | -0.005 |
Qu | -0.004 |
ora | -0.004 |
bl | -0.004 |
Token | Sleep |
Token ID | 17376 |
Feature activation | -6.41e+00 |
Pos logit contributions | |
supposed | +0.013 |
adders | +0.013 |
neys | +0.011 |
online | +0.011 |
about | +0.011 |
Neg logit contributions | |
maid | -0.016 |
fish | -0.015 |
Qu | -0.014 |
ora | -0.014 |
bl | -0.014 |
Token | well |
Token ID | 880 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
maid | +0.312 |
Qu | +0.294 |
snail | +0.293 |
fish | +0.292 |
ora | +0.282 |
Neg logit contributions | |
supposed | -0.181 |
about | -0.178 |
done | -0.166 |
adders | -0.154 |
late | -0.153 |
Token | ." |
Token ID | 526 |
Feature activation | -7.50e-01 |
Pos logit contributions | |
maid | +0.109 |
fish | +0.105 |
Qu | +0.098 |
ora | +0.098 |
bl | +0.097 |
Neg logit contributions | |
supposed | -0.090 |
adders | -0.090 |
neys | -0.077 |
By | -0.074 |
online | -0.074 |
Token |
|
Token ID | 198 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
maid | +0.036 |
fish | +0.034 |
Qu | +0.033 |
ora | +0.033 |
bl | +0.033 |
Neg logit contributions | |
supposed | -0.026 |
adders | -0.025 |
neys | -0.021 |
measuring | -0.020 |
online | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
maid | +0.002 |
fish | +0.002 |
Qu | +0.002 |
ora | +0.002 |
bl | +0.002 |
Neg logit contributions | |
supposed | -0.001 |
adders | -0.001 |
neys | -0.001 |
online | -0.001 |
measuring | -0.001 |
Token | The |
Token ID | 464 |
Feature activation | +2.70e-01 |
Pos logit contributions | |
maid | +0.001 |
fish | +0.001 |
breath | +0.001 |
bl | +0.001 |
ora | +0.001 |
Neg logit contributions | |
supposed | -0.001 |
adders | -0.001 |
neys | -0.000 |
By | -0.000 |
about | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | -3.45e-01 |
Pos logit contributions | |
adders | +0.007 |
supposed | +0.006 |
neys | +0.006 |
eps | +0.006 |
online | +0.005 |
Neg logit contributions | |
maid | -0.005 |
fish | -0.004 |
bl | -0.004 |
breath | -0.004 |
ora | -0.004 |
Token | my |
Token ID | 616 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
maid | +0.015 |
fish | +0.013 |
Qu | +0.013 |
ora | +0.013 |
bl | +0.013 |
Neg logit contributions | |
supposed | -0.013 |
adders | -0.013 |
neys | -0.012 |
online | -0.011 |
measuring | -0.011 |
Token | sweet |
Token ID | 6029 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
maid | +0.061 |
fish | +0.057 |
Qu | +0.055 |
ora | +0.054 |
bl | +0.053 |
Neg logit contributions | |
supposed | -0.053 |
adders | -0.052 |
neys | -0.046 |
about | -0.045 |
online | -0.044 |
Token | ies |
Token ID | 444 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
maid | +0.052 |
fish | +0.049 |
Qu | +0.049 |
ora | +0.043 |
breath | +0.043 |
Neg logit contributions | |
supposed | -0.077 |
adders | -0.073 |
about | -0.071 |
done | -0.068 |
online | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
adders | +0.008 |
supposed | +0.008 |
neys | +0.007 |
By | +0.007 |
online | +0.007 |
Neg logit contributions | |
maid | -0.010 |
fish | -0.010 |
bl | -0.009 |
ora | -0.009 |
Qu | -0.009 |
Token | Sleep |
Token ID | 17376 |
Feature activation | -6.40e+00 |
Pos logit contributions | |
maid | +0.001 |
Qu | +0.001 |
fish | +0.001 |
ora | +0.001 |
dolphin | +0.001 |
Neg logit contributions | |
supposed | -0.001 |
adders | -0.001 |
done | -0.001 |
neys | -0.001 |
online | -0.001 |
Token | tight |
Token ID | 5381 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
maid | +0.300 |
Qu | +0.286 |
fish | +0.281 |
snail | +0.275 |
jugg | +0.272 |
Neg logit contributions | |
supposed | -0.197 |
about | -0.188 |
done | -0.179 |
adders | -0.166 |
late | -0.164 |
Token | . |
Token ID | 13 |
Feature activation | +2.67e-03 |
Pos logit contributions | |
maid | +0.046 |
fish | +0.045 |
bl | +0.042 |
Qu | +0.042 |
ora | +0.042 |
Neg logit contributions | |
supposed | -0.036 |
adders | -0.035 |
neys | -0.030 |
By | -0.029 |
online | -0.029 |
Token | And |
Token ID | 843 |
Feature activation | +7.66e-01 |
Pos logit contributions | |
supposed | +0.000 |
adders | +0.000 |
neys | +0.000 |
online | +0.000 |
done | +0.000 |
Neg logit contributions | |
maid | -0.000 |
fish | -0.000 |
Qu | -0.000 |
ora | -0.000 |
bl | -0.000 |
Token | don |
Token ID | 836 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
supposed | +0.031 |
adders | +0.031 |
neys | +0.027 |
online | +0.026 |
By | +0.026 |
Neg logit contributions | |
maid | -0.031 |
fish | -0.029 |
Qu | -0.027 |
ora | -0.027 |
bl | -0.027 |
Token | 't |
Token ID | 470 |
Feature activation | -6.35e-01 |
Pos logit contributions | |
supposed | +0.005 |
adders | +0.005 |
neys | +0.005 |
online | +0.005 |
measuring | +0.005 |
Neg logit contributions | |
maid | -0.005 |
fish | -0.005 |
Qu | -0.005 |
bl | -0.005 |
breath | -0.005 |
Token | Good |
Token ID | 10248 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
maid | +0.072 |
fish | +0.066 |
breath | +0.063 |
ora | +0.062 |
dolphin | +0.058 |
Neg logit contributions | |
supposed | -0.041 |
late | -0.040 |
adders | -0.037 |
By | -0.036 |
� | -0.036 |
Token | night |
Token ID | 1755 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
maid | +0.053 |
fish | +0.051 |
Qu | +0.049 |
ora | +0.048 |
snail | +0.047 |
Neg logit contributions | |
supposed | -0.030 |
adders | -0.028 |
done | -0.026 |
about | -0.025 |
measuring | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
adders | +0.007 |
supposed | +0.007 |
neys | +0.006 |
online | +0.006 |
eps | +0.006 |
Neg logit contributions | |
maid | -0.007 |
fish | -0.007 |
breath | -0.007 |
bl | -0.007 |
Qu | -0.007 |
Token | doll |
Token ID | 3654 |
Feature activation | -3.94e-01 |
Pos logit contributions | |
maid | +0.009 |
fish | +0.009 |
Qu | +0.008 |
ora | +0.008 |
bl | +0.008 |
Neg logit contributions | |
supposed | -0.009 |
adders | -0.009 |
neys | -0.008 |
online | -0.007 |
By | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +2.68e-01 |
Pos logit contributions | |
maid | +0.018 |
fish | +0.017 |
Qu | +0.016 |
bl | +0.016 |
ora | +0.016 |
Neg logit contributions | |
supposed | -0.014 |
adders | -0.014 |
neys | -0.012 |
By | -0.012 |
online | -0.012 |
Token | Sleep |
Token ID | 17376 |
Feature activation | -6.39e+00 |
Pos logit contributions | |
supposed | +0.008 |
adders | +0.008 |
By | +0.007 |
about | +0.007 |
neys | +0.007 |
Neg logit contributions | |
maid | -0.013 |
fish | -0.013 |
ora | -0.012 |
breath | -0.012 |
bl | -0.011 |
Token | well |
Token ID | 880 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
maid | +0.331 |
Qu | +0.317 |
jugg | +0.316 |
fish | +0.316 |
snail | +0.315 |
Neg logit contributions | |
about | -0.164 |
supposed | -0.157 |
done | -0.150 |
late | -0.139 |
much | -0.132 |
Token | ." |
Token ID | 526 |
Feature activation | -8.61e-01 |
Pos logit contributions | |
maid | +0.116 |
fish | +0.110 |
Qu | +0.106 |
ora | +0.105 |
bl | +0.102 |
Neg logit contributions | |
supposed | -0.086 |
adders | -0.085 |
neys | -0.072 |
about | -0.071 |
online | -0.070 |
Token | Lily |
Token ID | 20037 |
Feature activation | -4.18e-01 |
Pos logit contributions | |
maid | +0.045 |
fish | +0.042 |
Qu | +0.041 |
ora | +0.041 |
bl | +0.041 |
Neg logit contributions | |
supposed | -0.025 |
adders | -0.025 |
neys | -0.020 |
measuring | -0.019 |
online | -0.019 |
Token | was |
Token ID | 373 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
maid | +0.018 |
fish | +0.017 |
ora | +0.017 |
Qu | +0.017 |
snail | +0.016 |
Neg logit contributions | |
about | -0.013 |
done | -0.013 |
supposed | -0.013 |
counted | -0.013 |
adders | -0.012 |
Token | happy |
Token ID | 3772 |
Feature activation | +6.91e-01 |
Pos logit contributions | |
adders | +0.011 |
supposed | +0.010 |
neys | +0.009 |
online | +0.009 |
By | +0.008 |
Neg logit contributions | |
maid | -0.015 |
bl | -0.013 |
fish | -0.013 |
Qu | -0.013 |
breath | -0.013 |
Token | you |
Token ID | 345 |
Feature activation | +8.28e-01 |
Pos logit contributions | |
maid | +0.006 |
fish | +0.005 |
Qu | +0.005 |
ora | +0.005 |
bl | +0.005 |
Neg logit contributions | |
supposed | -0.006 |
adders | -0.006 |
neys | -0.005 |
about | -0.005 |
online | -0.005 |
Token | need |
Token ID | 761 |
Feature activation | -5.59e-01 |
Pos logit contributions | |
supposed | +0.029 |
adders | +0.028 |
about | +0.024 |
done | +0.024 |
neys | +0.024 |
Neg logit contributions | |
maid | -0.039 |
fish | -0.037 |
ora | -0.036 |
Qu | -0.035 |
bl | -0.033 |
Token | to |
Token ID | 284 |
Feature activation | -5.87e-01 |
Pos logit contributions | |
maid | +0.031 |
fish | +0.029 |
Qu | +0.029 |
bl | +0.028 |
ora | +0.028 |
Neg logit contributions | |
supposed | -0.014 |
adders | -0.014 |
neys | -0.011 |
online | -0.011 |
By | -0.010 |
Token | sleep |
Token ID | 3993 |
Feature activation | -2.71e+00 |
Pos logit contributions | |
maid | +0.026 |
fish | +0.025 |
ora | +0.023 |
Qu | +0.023 |
dolphin | +0.022 |
Neg logit contributions | |
supposed | -0.022 |
adders | -0.021 |
done | -0.019 |
about | -0.019 |
measuring | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +3.94e-01 |
Pos logit contributions | |
maid | +0.126 |
fish | +0.120 |
Qu | +0.114 |
bl | +0.113 |
ora | +0.113 |
Neg logit contributions | |
supposed | -0.091 |
adders | -0.090 |
neys | -0.077 |
measuring | -0.073 |
By | -0.073 |
Token | Sleep |
Token ID | 17376 |
Feature activation | -6.38e+00 |
Pos logit contributions | |
supposed | +0.014 |
adders | +0.014 |
neys | +0.012 |
online | +0.011 |
about | +0.011 |
Neg logit contributions | |
maid | -0.018 |
fish | -0.017 |
Qu | -0.016 |
ora | -0.016 |
bl | -0.016 |
Token | will |
Token ID | 481 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
maid | +0.285 |
Qu | +0.269 |
fish | +0.258 |
jugg | +0.257 |
dolphin | +0.252 |
Neg logit contributions | |
supposed | -0.221 |
about | -0.216 |
done | -0.207 |
late | -0.193 |
adders | -0.186 |
Token | help |
Token ID | 1037 |
Feature activation | -3.55e+00 |
Pos logit contributions | |
maid | +0.054 |
fish | +0.049 |
ora | +0.048 |
Qu | +0.046 |
dolphin | +0.045 |
Neg logit contributions | |
supposed | -0.039 |
adders | -0.036 |
done | -0.034 |
about | -0.033 |
By | -0.032 |
Token | you |
Token ID | 345 |
Feature activation | +8.76e-01 |
Pos logit contributions | |
maid | +0.143 |
fish | +0.131 |
ora | +0.124 |
Qu | +0.123 |
breath | +0.116 |
Neg logit contributions | |
supposed | -0.148 |
adders | -0.142 |
about | -0.131 |
neys | -0.126 |
done | -0.126 |
Token | feel |
Token ID | 1254 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
supposed | +0.035 |
adders | +0.034 |
neys | +0.030 |
about | +0.030 |
online | +0.029 |
Neg logit contributions | |
maid | -0.036 |
fish | -0.033 |
Qu | -0.032 |
ora | -0.032 |
bl | -0.031 |
Token | better |
Token ID | 1365 |
Feature activation | -8.64e-02 |
Pos logit contributions | |
maid | +0.145 |
fish | +0.139 |
Qu | +0.134 |
ora | +0.133 |
bl | +0.131 |
Neg logit contributions | |
supposed | -0.065 |
adders | -0.064 |
neys | -0.051 |
about | -0.048 |
online | -0.048 |
Token | L |
Token ID | 43 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
supposed | +0.005 |
adders | +0.005 |
neys | +0.005 |
about | +0.004 |
online | +0.004 |
Neg logit contributions | |
maid | -0.006 |
fish | -0.005 |
Qu | -0.005 |
ora | -0.005 |
bl | -0.005 |
Token | ily |
Token ID | 813 |
Feature activation | -9.14e-01 |
Pos logit contributions | |
maid | +0.059 |
fish | +0.056 |
Qu | +0.054 |
ora | +0.054 |
bl | +0.053 |
Neg logit contributions | |
supposed | -0.039 |
adders | -0.038 |
neys | -0.033 |
online | -0.031 |
about | -0.031 |
Token | took |
Token ID | 1718 |
Feature activation | -8.64e-01 |
Pos logit contributions | |
maid | +0.038 |
fish | +0.036 |
Qu | +0.034 |
ora | +0.034 |
bl | +0.034 |
Neg logit contributions | |
supposed | -0.035 |
adders | -0.035 |
neys | -0.030 |
online | -0.030 |
about | -0.029 |
Token | a |
Token ID | 257 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
maid | +0.037 |
breath | +0.036 |
bl | +0.036 |
ora | +0.034 |
dolphin | +0.033 |
Neg logit contributions | |
supposed | -0.032 |
adders | -0.031 |
neys | -0.030 |
By | -0.026 |
measuring | -0.025 |
Token | deep |
Token ID | 2769 |
Feature activation | -8.63e-01 |
Pos logit contributions | |
maid | +0.042 |
fish | +0.040 |
Qu | +0.037 |
ora | +0.037 |
bl | +0.036 |
Neg logit contributions | |
supposed | -0.045 |
adders | -0.045 |
neys | -0.040 |
By | -0.038 |
about | -0.038 |
Token | breath |
Token ID | 8033 |
Feature activation | -6.36e+00 |
Pos logit contributions | |
maid | +0.018 |
Qu | +0.016 |
fish | +0.014 |
ora | +0.014 |
N | +0.011 |
Neg logit contributions | |
measuring | -0.047 |
about | -0.046 |
adders | -0.046 |
done | -0.046 |
outside | -0.045 |
Token | and |
Token ID | 290 |
Feature activation | +6.38e-02 |
Pos logit contributions | |
maid | +0.252 |
fish | +0.248 |
bl | +0.233 |
ora | +0.228 |
breath | +0.227 |
Neg logit contributions | |
adders | -0.254 |
supposed | -0.253 |
neys | -0.233 |
emergencies | -0.221 |
measuring | -0.205 |
Token | closed |
Token ID | 4838 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
supposed | +0.003 |
adders | +0.003 |
neys | +0.002 |
By | +0.002 |
online | +0.002 |
Neg logit contributions | |
maid | -0.002 |
fish | -0.002 |
Qu | -0.002 |
ora | -0.002 |
bl | -0.002 |
Token | her |
Token ID | 607 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
maid | +0.089 |
fish | +0.085 |
Qu | +0.082 |
ora | +0.080 |
bl | +0.076 |
Neg logit contributions | |
supposed | -0.068 |
adders | -0.068 |
about | -0.059 |
online | -0.058 |
measuring | -0.057 |
Token | eyes |
Token ID | 2951 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
maid | +0.031 |
fish | +0.028 |
Qu | +0.027 |
ora | +0.026 |
bl | +0.024 |
Neg logit contributions | |
supposed | -0.055 |
adders | -0.055 |
neys | -0.050 |
about | -0.049 |
online | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | +8.00e-03 |
Pos logit contributions | |
maid | +0.062 |
fish | +0.059 |
bl | +0.059 |
breath | +0.056 |
ora | +0.054 |
Neg logit contributions | |
supposed | -0.064 |
adders | -0.060 |
neys | -0.056 |
emergencies | -0.054 |
elev | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
supposed | +0.006 |
adders | +0.006 |
neys | +0.005 |
By | +0.005 |
emergencies | +0.005 |
Neg logit contributions | |
maid | -0.010 |
fish | -0.009 |
Qu | -0.009 |
bl | -0.009 |
ora | -0.009 |
Token | It |
Token ID | 632 |
Feature activation | -6.28e-01 |
Pos logit contributions | |
supposed | +0.001 |
adders | +0.001 |
neys | +0.001 |
done | +0.001 |
emergencies | +0.001 |
Neg logit contributions | |
Qu | -0.001 |
maid | -0.001 |
ora | -0.001 |
bl | -0.001 |
fish | -0.001 |
Token | took |
Token ID | 1718 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
maid | +0.030 |
ora | +0.026 |
fish | +0.026 |
Qu | +0.026 |
dolphin | +0.025 |
Neg logit contributions | |
supposed | -0.021 |
adders | -0.021 |
about | -0.020 |
done | -0.019 |
neys | -0.019 |
Token | a |
Token ID | 257 |
Feature activation | -7.92e-01 |
Pos logit contributions | |
maid | +0.049 |
breath | +0.046 |
fish | +0.046 |
bl | +0.046 |
Qu | +0.046 |
Neg logit contributions | |
supposed | -0.036 |
adders | -0.034 |
neys | -0.030 |
online | -0.028 |
By | -0.028 |
Token | deep |
Token ID | 2769 |
Feature activation | -8.33e-01 |
Pos logit contributions | |
maid | +0.037 |
fish | +0.034 |
Qu | +0.034 |
ora | +0.034 |
bl | +0.032 |
Neg logit contributions | |
supposed | -0.027 |
adders | -0.026 |
about | -0.022 |
measuring | -0.022 |
online | -0.022 |
Token | breath |
Token ID | 8033 |
Feature activation | -6.36e+00 |
Pos logit contributions | |
maid | +0.026 |
Qu | +0.025 |
fish | +0.022 |
ora | +0.022 |
dolphin | +0.021 |
Neg logit contributions | |
about | -0.038 |
outside | -0.037 |
done | -0.037 |
supposed | -0.036 |
inside | -0.036 |
Token | and |
Token ID | 290 |
Feature activation | -4.64e-02 |
Pos logit contributions | |
maid | +0.278 |
fish | +0.269 |
Qu | +0.250 |
ora | +0.250 |
bl | +0.249 |
Neg logit contributions | |
supposed | -0.230 |
adders | -0.229 |
neys | -0.202 |
emergencies | -0.192 |
measuring | -0.187 |
Token | relaxed |
Token ID | 18397 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
maid | +0.002 |
fish | +0.002 |
Qu | +0.002 |
ora | +0.002 |
bl | +0.002 |
Neg logit contributions | |
supposed | -0.002 |
adders | -0.002 |
about | -0.002 |
neys | -0.002 |
online | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
maid | +0.103 |
fish | +0.096 |
Qu | +0.095 |
ora | +0.094 |
bl | +0.090 |
Neg logit contributions | |
supposed | -0.045 |
adders | -0.044 |
about | -0.040 |
done | -0.036 |
online | -0.035 |
Token | |
Token ID | 220 |
Feature activation | -6.88e-01 |
Pos logit contributions | |
Qu | +0.006 |
maid | +0.006 |
ora | +0.006 |
fish | +0.005 |
bl | +0.005 |
Neg logit contributions | |
supposed | -0.006 |
adders | -0.005 |
neys | -0.005 |
done | -0.005 |
emergencies | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +5.72e-01 |
Pos logit contributions | |
maid | +0.024 |
ora | +0.023 |
fish | +0.023 |
Qu | +0.023 |
of | +0.022 |
Neg logit contributions | |
supposed | -0.028 |
adders | -0.026 |
emergencies | -0.024 |
online | -0.024 |
measuring | -0.024 |
Token | Good |
Token ID | 10248 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
maid | +0.040 |
fish | +0.037 |
ora | +0.035 |
breath | +0.034 |
bl | +0.034 |
Neg logit contributions | |
supposed | -0.029 |
adders | -0.028 |
By | -0.025 |
neys | -0.024 |
about | -0.024 |
Token | night |
Token ID | 3847 |
Feature activation | -3.52e+00 |
Pos logit contributions | |
maid | +0.049 |
fish | +0.047 |
Qu | +0.046 |
ora | +0.044 |
snail | +0.042 |
Neg logit contributions | |
supposed | -0.033 |
adders | -0.029 |
about | -0.029 |
done | -0.028 |
driving | -0.026 |
Token | , |
Token ID | 11 |
Feature activation | -2.36e-01 |
Pos logit contributions | |
maid | +0.164 |
fish | +0.155 |
Qu | +0.146 |
ora | +0.145 |
bl | +0.145 |
Neg logit contributions | |
adders | -0.123 |
supposed | -0.123 |
neys | -0.106 |
online | -0.100 |
By | -0.098 |
Token | Max |
Token ID | 5436 |
Feature activation | -9.89e-01 |
Pos logit contributions | |
maid | +0.013 |
fish | +0.013 |
ora | +0.012 |
Qu | +0.012 |
bl | +0.012 |
Neg logit contributions | |
supposed | -0.006 |
adders | -0.006 |
about | -0.005 |
neys | -0.004 |
online | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
maid | +0.045 |
fish | +0.043 |
Qu | +0.041 |
ora | +0.041 |
bl | +0.040 |
Neg logit contributions | |
supposed | -0.035 |
adders | -0.034 |
neys | -0.029 |
about | -0.029 |
online | -0.028 |
Token | Sleep |
Token ID | 17376 |
Feature activation | -6.35e+00 |
Pos logit contributions | |
supposed | +0.010 |
adders | +0.010 |
neys | +0.009 |
By | +0.009 |
about | +0.009 |
Neg logit contributions | |
maid | -0.013 |
fish | -0.013 |
ora | -0.012 |
Qu | -0.012 |
bl | -0.012 |
Token | tight |
Token ID | 5381 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
maid | +0.305 |
Qu | +0.293 |
snail | +0.289 |
fish | +0.285 |
jugg | +0.281 |
Neg logit contributions | |
about | -0.184 |
supposed | -0.182 |
done | -0.170 |
late | -0.161 |
much | -0.143 |
Token | ." |
Token ID | 526 |
Feature activation | -7.26e-01 |
Pos logit contributions | |
maid | +0.052 |
ora | +0.048 |
fish | +0.047 |
Qu | +0.047 |
dolphin | +0.045 |
Neg logit contributions | |
supposed | -0.035 |
adders | -0.034 |
about | -0.031 |
done | -0.030 |
driving | -0.029 |
Token | Max |
Token ID | 5436 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
maid | +0.034 |
Qu | +0.033 |
fish | +0.032 |
bl | +0.032 |
breath | +0.031 |
Neg logit contributions | |
adders | -0.024 |
supposed | -0.024 |
neys | -0.019 |
measuring | -0.018 |
online | -0.018 |
Token | fell |
Token ID | 3214 |
Feature activation | +4.48e-01 |
Pos logit contributions | |
maid | +0.042 |
ora | +0.038 |
fish | +0.037 |
Qu | +0.036 |
snail | +0.036 |
Neg logit contributions | |
about | -0.035 |
done | -0.035 |
� | -0.034 |
supposed | -0.033 |
counted | -0.033 |
Token | asleep |
Token ID | 16039 |
Feature activation | -2.75e+00 |
Pos logit contributions | |
supposed | +0.016 |
adders | +0.016 |
about | +0.014 |
driving | +0.014 |
online | +0.014 |
Neg logit contributions | |
maid | -0.020 |
fish | -0.019 |
ora | -0.018 |
Qu | -0.018 |
dolphin | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | -4.76e-01 |
Pos logit contributions | |
adders | +0.005 |
supposed | +0.005 |
neys | +0.004 |
online | +0.004 |
eps | +0.004 |
Neg logit contributions | |
maid | -0.005 |
fish | -0.005 |
breath | -0.004 |
ora | -0.004 |
bl | -0.004 |
Token | Anna |
Token ID | 11735 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
maid | +0.020 |
fish | +0.019 |
Qu | +0.019 |
ora | +0.018 |
bl | +0.018 |
Neg logit contributions | |
supposed | -0.018 |
adders | -0.018 |
about | -0.015 |
neys | -0.015 |
online | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | -9.17e-02 |
Pos logit contributions | |
maid | +0.054 |
fish | +0.052 |
bl | +0.049 |
ora | +0.048 |
Qu | +0.047 |
Neg logit contributions | |
adders | -0.059 |
supposed | -0.057 |
neys | -0.050 |
elev | -0.048 |
online | -0.048 |
Token | Ben |
Token ID | 3932 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
maid | +0.003 |
fish | +0.003 |
Qu | +0.003 |
ora | +0.003 |
bl | +0.003 |
Neg logit contributions | |
supposed | -0.004 |
adders | -0.004 |
neys | -0.004 |
about | -0.004 |
online | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -6.90e-02 |
Pos logit contributions | |
maid | +0.104 |
fish | +0.100 |
bl | +0.098 |
ora | +0.093 |
Qu | +0.092 |
Neg logit contributions | |
adders | -0.099 |
supposed | -0.094 |
neys | -0.082 |
By | -0.079 |
elev | -0.078 |
Token | Sleep |
Token ID | 17376 |
Feature activation | -6.34e+00 |
Pos logit contributions | |
maid | +0.003 |
fish | +0.003 |
Qu | +0.003 |
ora | +0.003 |
bl | +0.003 |
Neg logit contributions | |
supposed | -0.002 |
adders | -0.002 |
neys | -0.002 |
online | -0.002 |
done | -0.002 |
Token | well |
Token ID | 880 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
maid | +0.306 |
Qu | +0.292 |
snail | +0.287 |
fish | +0.285 |
jugg | +0.284 |
Neg logit contributions | |
supposed | -0.183 |
about | -0.177 |
done | -0.170 |
late | -0.157 |
neys | -0.150 |
Token | . |
Token ID | 13 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
maid | +0.106 |
fish | +0.106 |
bl | +0.097 |
ora | +0.097 |
Qu | +0.095 |
Neg logit contributions | |
supposed | -0.082 |
adders | -0.081 |
neys | -0.068 |
By | -0.067 |
emergencies | -0.066 |
Token | We |
Token ID | 775 |
Feature activation | +9.40e-01 |
Pos logit contributions | |
maid | +0.001 |
fish | +0.001 |
Qu | +0.001 |
ora | +0.001 |
bl | +0.001 |
Neg logit contributions | |
supposed | -0.001 |
adders | -0.001 |
neys | -0.001 |
about | -0.001 |
By | -0.001 |
Token | love |
Token ID | 1842 |
Feature activation | -6.84e-01 |
Pos logit contributions | |
supposed | +0.034 |
adders | +0.033 |
neys | +0.028 |
about | +0.028 |
online | +0.028 |
Neg logit contributions | |
maid | -0.043 |
fish | -0.040 |
ora | -0.038 |
Qu | -0.038 |
bl | -0.037 |
Token | you |
Token ID | 345 |
Feature activation | +8.06e-01 |
Pos logit contributions | |
maid | +0.028 |
fish | +0.026 |
Qu | +0.025 |
ora | +0.025 |
bl | +0.024 |
Neg logit contributions | |
supposed | -0.028 |
adders | -0.028 |
neys | -0.024 |
about | -0.023 |
online | -0.023 |
Token | Sara |
Token ID | 24799 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
maid | +0.014 |
fish | +0.013 |
Qu | +0.013 |
ora | +0.012 |
breath | +0.012 |
Neg logit contributions | |
about | -0.009 |
supposed | -0.009 |
adders | -0.009 |
measuring | -0.008 |
online | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
maid | +0.016 |
fish | +0.015 |
bl | +0.015 |
ora | +0.014 |
Qu | +0.014 |
Neg logit contributions | |
adders | -0.014 |
supposed | -0.014 |
neys | -0.012 |
By | -0.012 |
online | -0.012 |
Token | Good |
Token ID | 4599 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
supposed | +0.007 |
adders | +0.007 |
neys | +0.006 |
done | +0.006 |
online | +0.006 |
Neg logit contributions | |
maid | -0.008 |
fish | -0.007 |
Qu | -0.007 |
ora | -0.007 |
bl | -0.007 |
Token | night |
Token ID | 1755 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
supposed | +0.030 |
adders | +0.026 |
done | +0.025 |
about | +0.024 |
clock | +0.022 |
Neg logit contributions | |
maid | -0.082 |
fish | -0.080 |
Qu | -0.079 |
ora | -0.076 |
snail | -0.075 |
Token | . |
Token ID | 13 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
adders | +0.011 |
supposed | +0.010 |
neys | +0.009 |
online | +0.009 |
eps | +0.009 |
Neg logit contributions | |
maid | -0.010 |
fish | -0.009 |
bl | -0.009 |
breath | -0.008 |
ora | -0.008 |
Token | Sleep |
Token ID | 17376 |
Feature activation | -6.34e+00 |
Pos logit contributions | |
supposed | +0.008 |
adders | +0.008 |
neys | +0.007 |
online | +0.006 |
about | +0.006 |
Neg logit contributions | |
maid | -0.010 |
fish | -0.009 |
Qu | -0.009 |
ora | -0.009 |
bl | -0.008 |
Token | tight |
Token ID | 5381 |
Feature activation | -9.12e-01 |
Pos logit contributions | |
maid | +0.298 |
Qu | +0.286 |
snail | +0.285 |
fish | +0.280 |
jugg | +0.275 |
Neg logit contributions | |
supposed | -0.183 |
about | -0.182 |
done | -0.169 |
late | -0.159 |
neys | -0.152 |
Token | . |
Token ID | 13 |
Feature activation | +2.37e-01 |
Pos logit contributions | |
maid | +0.043 |
fish | +0.041 |
Qu | +0.039 |
ora | +0.039 |
bl | +0.038 |
Neg logit contributions | |
supposed | -0.030 |
adders | -0.030 |
neys | -0.025 |
about | -0.025 |
online | -0.024 |
Token | Don |
Token ID | 2094 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
supposed | +0.008 |
adders | +0.008 |
neys | +0.007 |
online | +0.007 |
about | +0.007 |
Neg logit contributions | |
maid | -0.011 |
fish | -0.010 |
Qu | -0.010 |
ora | -0.010 |
bl | -0.010 |
Token | 't |
Token ID | 470 |
Feature activation | -5.88e-01 |
Pos logit contributions | |
supposed | +0.005 |
adders | +0.005 |
neys | +0.004 |
online | +0.004 |
measuring | +0.004 |
Neg logit contributions | |
maid | -0.005 |
fish | -0.005 |
Qu | -0.005 |
bl | -0.005 |
ora | -0.005 |
Token | let |
Token ID | 1309 |
Feature activation | -2.31e+00 |
Pos logit contributions | |
maid | +0.024 |
fish | +0.023 |
ora | +0.022 |
Qu | +0.021 |
dolphin | +0.020 |
Neg logit contributions | |
supposed | -0.023 |
adders | -0.021 |
about | -0.020 |
done | -0.020 |
measuring | -0.019 |
Token | Good |
Token ID | 10248 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
maid | +0.030 |
fish | +0.027 |
breath | +0.025 |
ora | +0.024 |
bl | +0.023 |
Neg logit contributions | |
supposed | -0.036 |
adders | -0.034 |
By | -0.032 |
late | -0.032 |
neys | -0.031 |
Token | night |
Token ID | 3847 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
maid | +0.052 |
fish | +0.051 |
Qu | +0.049 |
ora | +0.047 |
snail | +0.046 |
Neg logit contributions | |
supposed | -0.031 |
adders | -0.028 |
about | -0.025 |
done | -0.025 |
online | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
maid | +0.163 |
fish | +0.150 |
bl | +0.144 |
ora | +0.143 |
Qu | +0.142 |
Neg logit contributions | |
adders | -0.128 |
supposed | -0.125 |
neys | -0.111 |
online | -0.102 |
By | -0.100 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
maid | +0.006 |
fish | +0.006 |
ora | +0.006 |
Qu | +0.005 |
breath | +0.005 |
Neg logit contributions | |
supposed | -0.004 |
adders | -0.004 |
about | -0.004 |
done | -0.003 |
online | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
maid | +0.009 |
fish | +0.008 |
Qu | +0.008 |
ora | +0.008 |
bl | +0.008 |
Neg logit contributions | |
adders | -0.007 |
supposed | -0.007 |
neys | -0.006 |
online | -0.006 |
By | -0.006 |
Token | Sleep |
Token ID | 17376 |
Feature activation | -6.34e+00 |
Pos logit contributions | |
supposed | +0.007 |
adders | +0.007 |
neys | +0.006 |
online | +0.006 |
about | +0.006 |
Neg logit contributions | |
maid | -0.009 |
fish | -0.008 |
Qu | -0.008 |
ora | -0.008 |
bl | -0.008 |
Token | well |
Token ID | 880 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
maid | +0.316 |
fish | +0.303 |
Qu | +0.302 |
snail | +0.298 |
jugg | +0.295 |
Neg logit contributions | |
about | -0.184 |
supposed | -0.183 |
done | -0.166 |
late | -0.163 |
much | -0.144 |
Token | ." |
Token ID | 526 |
Feature activation | -7.87e-01 |
Pos logit contributions | |
maid | +0.114 |
fish | +0.109 |
Qu | +0.102 |
ora | +0.102 |
bl | +0.101 |
Neg logit contributions | |
supposed | -0.092 |
adders | -0.092 |
neys | -0.079 |
online | -0.076 |
By | -0.076 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
maid | +0.035 |
fish | +0.033 |
Qu | +0.032 |
ora | +0.032 |
bl | +0.031 |
Neg logit contributions | |
supposed | -0.029 |
adders | -0.028 |
neys | -0.024 |
about | -0.024 |
online | -0.024 |
Token | replied |
Token ID | 8712 |
Feature activation | +9.75e-02 |
Pos logit contributions | |
maid | +0.009 |
Qu | +0.008 |
ora | +0.008 |
fish | +0.008 |
snail | +0.008 |
Neg logit contributions | |
about | -0.007 |
done | -0.006 |
supposed | -0.006 |
counted | -0.006 |
� | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
supposed | +0.002 |
adders | +0.002 |
about | +0.002 |
measuring | +0.002 |
driving | +0.002 |
Neg logit contributions | |
maid | -0.005 |
ora | -0.005 |
fish | -0.005 |
Qu | -0.005 |
dolphin | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
maid | +0.037 |
fish | +0.036 |
breath | +0.034 |
bl | +0.034 |
ora | +0.032 |
Neg logit contributions | |
adders | -0.033 |
supposed | -0.032 |
neys | -0.028 |
emergencies | -0.027 |
elev | -0.026 |
Token | gave |
Token ID | 2921 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
supposed | +0.010 |
adders | +0.010 |
neys | +0.009 |
about | +0.009 |
online | +0.009 |
Neg logit contributions | |
maid | -0.011 |
fish | -0.010 |
Qu | -0.010 |
ora | -0.010 |
bl | -0.010 |
Token | it |
Token ID | 340 |
Feature activation | -6.30e-01 |
Pos logit contributions | |
maid | +0.046 |
fish | +0.043 |
Qu | +0.042 |
ora | +0.041 |
bl | +0.041 |
Neg logit contributions | |
supposed | -0.045 |
adders | -0.044 |
neys | -0.039 |
By | -0.038 |
online | -0.037 |
Token | a |
Token ID | 257 |
Feature activation | -6.17e-01 |
Pos logit contributions | |
breath | +0.028 |
maid | +0.028 |
bl | +0.027 |
fish | +0.027 |
Qu | +0.027 |
Neg logit contributions | |
supposed | -0.020 |
adders | -0.020 |
neys | -0.017 |
By | -0.017 |
emergencies | -0.017 |
Token | light |
Token ID | 1657 |
Feature activation | -6.16e-01 |
Pos logit contributions | |
maid | +0.029 |
fish | +0.027 |
ora | +0.026 |
Qu | +0.026 |
bl | +0.025 |
Neg logit contributions | |
supposed | -0.021 |
adders | -0.021 |
about | -0.018 |
neys | -0.018 |
online | -0.018 |
Token | burial |
Token ID | 23867 |
Feature activation | -6.33e+00 |
Pos logit contributions | |
maid | +0.031 |
fish | +0.030 |
bl | +0.028 |
Qu | +0.027 |
breath | +0.027 |
Neg logit contributions | |
supposed | -0.020 |
adders | -0.019 |
neys | -0.016 |
By | -0.016 |
emergencies | -0.015 |
Token | in |
Token ID | 287 |
Feature activation | -9.91e-01 |
Pos logit contributions | |
maid | +0.295 |
fish | +0.288 |
ora | +0.263 |
bl | +0.261 |
breath | +0.260 |
Neg logit contributions | |
supposed | -0.215 |
adders | -0.211 |
neys | -0.178 |
By | -0.177 |
online | -0.172 |
Token | the |
Token ID | 262 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
maid | +0.049 |
fish | +0.046 |
ora | +0.045 |
Qu | +0.044 |
bl | +0.043 |
Neg logit contributions | |
supposed | -0.031 |
adders | -0.031 |
neys | -0.026 |
about | -0.025 |
measuring | -0.025 |
Token | woods |
Token ID | 16479 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
supposed | +0.040 |
adders | +0.040 |
neys | +0.034 |
about | +0.033 |
online | +0.033 |
Neg logit contributions | |
maid | -0.047 |
fish | -0.045 |
Qu | -0.043 |
ora | -0.042 |
bl | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | +5.94e-02 |
Pos logit contributions | |
supposed | +0.061 |
adders | +0.059 |
emergencies | +0.051 |
neys | +0.050 |
eps | +0.050 |
Neg logit contributions | |
fish | -0.060 |
maid | -0.059 |
bl | -0.053 |
breath | -0.053 |
Qu | -0.050 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.36e+00 |
Pos logit contributions | |
supposed | +0.002 |
adders | +0.002 |
neys | +0.002 |
emergencies | +0.002 |
online | +0.002 |
Neg logit contributions | |
maid | -0.003 |
Qu | -0.003 |
fish | -0.002 |
ora | -0.002 |
bl | -0.002 |
Token | trees |
Token ID | 7150 |
Feature activation | +4.89e+00 |
Pos logit contributions | |
bl | +0.043 |
fish | +0.042 |
breath | +0.042 |
dolphin | +0.042 |
maid | +0.041 |
Neg logit contributions | |
supposed | -0.054 |
By | -0.045 |
adders | -0.045 |
neys | -0.042 |
ends | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
supposed | +0.230 |
adders | +0.218 |
emergencies | +0.202 |
neys | +0.200 |
By | +0.192 |
Neg logit contributions | |
fish | -0.149 |
bl | -0.147 |
breath | -0.146 |
maid | -0.132 |
snail | -0.126 |
Token | blew |
Token ID | 17948 |
Feature activation | -6.23e-01 |
Pos logit contributions | |
supposed | +0.010 |
adders | +0.009 |
neys | +0.008 |
By | +0.008 |
about | +0.008 |
Neg logit contributions | |
bl | -0.008 |
breath | -0.008 |
fish | -0.008 |
maid | -0.008 |
dolphin | -0.008 |
Token | away |
Token ID | 1497 |
Feature activation | -2.61e+00 |
Pos logit contributions | |
breath | +0.024 |
maid | +0.023 |
bl | +0.023 |
Qu | +0.022 |
fish | +0.022 |
Neg logit contributions | |
supposed | -0.027 |
adders | -0.024 |
neys | -0.022 |
By | -0.021 |
selling | -0.021 |
Token | some |
Token ID | 617 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
maid | +0.110 |
fish | +0.105 |
breath | +0.102 |
bl | +0.100 |
ora | +0.098 |
Neg logit contributions | |
supposed | -0.102 |
adders | -0.091 |
By | -0.080 |
online | -0.079 |
measuring | -0.079 |
Token | roofs |
Token ID | 42251 |
Feature activation | +5.00e+00 |
Pos logit contributions | |
breath | +0.040 |
fish | +0.039 |
bl | +0.039 |
maid | +0.037 |
dolphin | +0.036 |
Neg logit contributions | |
supposed | -0.053 |
adders | -0.047 |
By | -0.046 |
neys | -0.044 |
online | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
supposed | +0.207 |
adders | +0.205 |
emergencies | +0.191 |
By | +0.187 |
measuring | +0.179 |
Neg logit contributions | |
fish | -0.181 |
maid | -0.169 |
breath | -0.167 |
bl | -0.160 |
ora | -0.148 |
Token | But |
Token ID | 887 |
Feature activation | +7.95e-01 |
Pos logit contributions | |
supposed | +0.006 |
adders | +0.005 |
done | +0.005 |
eps | +0.005 |
neys | +0.004 |
Neg logit contributions | |
Qu | -0.005 |
maid | -0.005 |
bl | -0.005 |
ora | -0.005 |
dolphin | -0.005 |
Token | eventually |
Token ID | 4191 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
supposed | +0.035 |
adders | +0.033 |
neys | +0.031 |
chairs | +0.027 |
clock | +0.027 |
Neg logit contributions | |
fish | -0.029 |
maid | -0.029 |
breath | -0.027 |
Qu | -0.027 |
bl | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
supposed | +0.042 |
adders | +0.042 |
neys | +0.037 |
online | +0.034 |
By | +0.033 |
Neg logit contributions | |
maid | -0.056 |
fish | -0.053 |
Qu | -0.051 |
ora | -0.050 |
bl | -0.050 |
Token | it |
Token ID | 340 |
Feature activation | -5.56e-01 |
Pos logit contributions | |
fish | +0.005 |
breath | +0.005 |
maid | +0.005 |
dolphin | +0.005 |
Qu | +0.005 |
Neg logit contributions | |
supposed | -0.007 |
adders | -0.006 |
neys | -0.006 |
old | -0.006 |
chairs | -0.005 |
Token | wouldn |
Token ID | 3636 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
supposed | +0.027 |
adders | +0.027 |
neys | +0.024 |
about | +0.023 |
By | +0.023 |
Neg logit contributions | |
maid | -0.030 |
fish | -0.029 |
Qu | -0.027 |
ora | -0.027 |
bl | -0.026 |
Token | 't |
Token ID | 470 |
Feature activation | -5.60e-01 |
Pos logit contributions | |
supposed | +0.058 |
adders | +0.058 |
neys | +0.049 |
about | +0.048 |
online | +0.048 |
Neg logit contributions | |
maid | -0.079 |
fish | -0.074 |
Qu | -0.071 |
ora | -0.071 |
bl | -0.069 |
Token | get |
Token ID | 651 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
maid | +0.023 |
fish | +0.021 |
ora | +0.020 |
Qu | +0.020 |
bl | +0.019 |
Neg logit contributions | |
supposed | -0.023 |
adders | -0.022 |
about | -0.020 |
done | -0.019 |
neys | -0.019 |
Token | stuck |
Token ID | 7819 |
Feature activation | +4.14e-01 |
Pos logit contributions | |
maid | +0.013 |
fish | +0.013 |
bl | +0.012 |
breath | +0.012 |
Qu | +0.012 |
Neg logit contributions | |
adders | -0.013 |
supposed | -0.012 |
neys | -0.011 |
By | -0.011 |
online | -0.011 |
Token | in |
Token ID | 287 |
Feature activation | -9.69e-01 |
Pos logit contributions | |
adders | +0.015 |
supposed | +0.015 |
neys | +0.014 |
By | +0.013 |
measuring | +0.012 |
Neg logit contributions | |
maid | -0.018 |
fish | -0.017 |
breath | -0.016 |
bl | -0.016 |
Qu | -0.016 |
Token | trees |
Token ID | 7150 |
Feature activation | +5.00e+00 |
Pos logit contributions | |
maid | +0.048 |
fish | +0.046 |
Qu | +0.044 |
ora | +0.044 |
bl | +0.043 |
Neg logit contributions | |
supposed | -0.030 |
adders | -0.029 |
neys | -0.025 |
about | -0.024 |
online | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
supposed | +0.212 |
adders | +0.208 |
By | +0.185 |
emergencies | +0.183 |
neys | +0.183 |
Neg logit contributions | |
fish | -0.184 |
maid | -0.173 |
bl | -0.166 |
breath | -0.158 |
Qu | -0.151 |
Token | And |
Token ID | 843 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
adders | +0.012 |
supposed | +0.012 |
neys | +0.010 |
measuring | +0.009 |
eps | +0.009 |
Neg logit contributions | |
maid | -0.013 |
ora | -0.013 |
Qu | -0.013 |
bl | -0.012 |
fish | -0.012 |
Token | she |
Token ID | 673 |
Feature activation | +4.65e-01 |
Pos logit contributions | |
adders | +0.038 |
supposed | +0.038 |
neys | +0.032 |
online | +0.030 |
measuring | +0.030 |
Neg logit contributions | |
maid | -0.056 |
fish | -0.053 |
Qu | -0.051 |
ora | -0.050 |
bl | -0.050 |
Token | was |
Token ID | 373 |
Feature activation | +7.05e-01 |
Pos logit contributions | |
supposed | +0.015 |
adders | +0.014 |
done | +0.012 |
about | +0.012 |
neys | +0.012 |
Neg logit contributions | |
maid | -0.023 |
fish | -0.021 |
ora | -0.021 |
Qu | -0.021 |
breath | -0.021 |
Token | always |
Token ID | 1464 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
adders | +0.026 |
neys | +0.023 |
cases | +0.022 |
emergencies | +0.022 |
supposed | +0.022 |
Neg logit contributions | |
bl | -0.029 |
maid | -0.029 |
breath | -0.027 |
fish | -0.027 |
Qu | -0.027 |
Token | liked |
Token ID | 8288 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
maid | +0.013 |
fish | +0.012 |
Qu | +0.012 |
ora | +0.012 |
bl | +0.012 |
Neg logit contributions | |
supposed | -0.013 |
adders | -0.013 |
neys | -0.011 |
about | -0.011 |
online | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | -7.11e-01 |
Pos logit contributions | |
fish | +0.024 |
maid | +0.023 |
dolphin | +0.023 |
bl | +0.023 |
breath | +0.023 |
Neg logit contributions | |
adders | -0.017 |
supposed | -0.016 |
By | -0.014 |
neys | -0.014 |
Until | -0.013 |
Token | pretend |
Token ID | 16614 |
Feature activation | -8.13e-01 |
Pos logit contributions | |
maid | +0.030 |
fish | +0.027 |
Qu | +0.026 |
ora | +0.026 |
bl | +0.025 |
Neg logit contributions | |
supposed | -0.028 |
adders | -0.028 |
measuring | -0.024 |
about | -0.024 |
online | -0.024 |
Token | they |
Token ID | 484 |
Feature activation | +3.85e-01 |
Pos logit contributions | |
fish | +0.033 |
maid | +0.031 |
bl | +0.029 |
dolphin | +0.029 |
breath | +0.027 |
Neg logit contributions | |
supposed | -0.034 |
adders | -0.032 |
By | -0.030 |
Until | -0.029 |
neys | -0.028 |
Token | were |
Token ID | 547 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
adders | +0.012 |
supposed | +0.011 |
neys | +0.010 |
about | +0.010 |
online | +0.010 |
Neg logit contributions | |
maid | -0.020 |
fish | -0.018 |
Qu | -0.017 |
ora | -0.017 |
breath | -0.017 |
Token | lions |
Token ID | 31420 |
Feature activation | +5.30e+00 |
Pos logit contributions | |
By | +0.063 |
chairs | +0.061 |
supposed | +0.061 |
adders | +0.060 |
Until | +0.059 |
Neg logit contributions | |
maid | -0.057 |
fish | -0.056 |
mer | -0.054 |
dolphin | -0.054 |
ora | -0.052 |
Token | and |
Token ID | 290 |
Feature activation | +6.83e-02 |
Pos logit contributions | |
supposed | +0.215 |
adders | +0.200 |
By | +0.179 |
neys | +0.175 |
online | +0.172 |
Neg logit contributions | |
fish | -0.221 |
maid | -0.216 |
bl | -0.196 |
Qu | -0.193 |
lizard | -0.191 |
Token | tigers |
Token ID | 43114 |
Feature activation | +5.39e+00 |
Pos logit contributions | |
supposed | +0.002 |
adders | +0.002 |
neys | +0.002 |
online | +0.002 |
By | +0.002 |
Neg logit contributions | |
maid | -0.003 |
fish | -0.003 |
Qu | -0.003 |
bl | -0.003 |
ora | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -3.00e-01 |
Pos logit contributions | |
supposed | +0.203 |
adders | +0.195 |
By | +0.168 |
neys | +0.167 |
online | +0.165 |
Neg logit contributions | |
fish | -0.237 |
maid | -0.229 |
Qu | -0.212 |
bl | -0.207 |
ora | -0.206 |
Token | They |
Token ID | 1119 |
Feature activation | -3.77e-01 |
Pos logit contributions | |
maid | +0.013 |
Qu | +0.013 |
ora | +0.013 |
fish | +0.012 |
dolphin | +0.012 |
Neg logit contributions | |
supposed | -0.012 |
adders | -0.010 |
fixing | -0.009 |
driving | -0.009 |
neys | -0.009 |
Token | would |
Token ID | 561 |
Feature activation | -3.53e-01 |
Pos logit contributions | |
maid | +0.016 |
fish | +0.015 |
Qu | +0.015 |
ora | +0.014 |
bl | +0.014 |
Neg logit contributions | |
supposed | -0.015 |
adders | -0.014 |
neys | -0.012 |
By | -0.012 |
emergencies | -0.012 |
Token | only |
Token ID | 691 |
Feature activation | -6.31e-02 |
Pos logit contributions | |
supposed | +0.023 |
adders | +0.022 |
By | +0.019 |
online | +0.019 |
neys | +0.019 |
Neg logit contributions | |
maid | -0.028 |
fish | -0.028 |
bl | -0.027 |
breath | -0.026 |
lizard | -0.026 |
Token | grass |
Token ID | 8701 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
maid | +0.003 |
fish | +0.002 |
bl | +0.002 |
mer | +0.002 |
lizard | +0.002 |
Neg logit contributions | |
supposed | -0.002 |
adders | -0.002 |
By | -0.002 |
neys | -0.002 |
online | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | +2.83e-01 |
Pos logit contributions | |
supposed | +0.053 |
adders | +0.050 |
By | +0.044 |
neys | +0.043 |
emergencies | +0.043 |
Neg logit contributions | |
maid | -0.058 |
fish | -0.058 |
bl | -0.051 |
ora | -0.051 |
Qu | -0.050 |
Token | rocks |
Token ID | 12586 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
supposed | +0.012 |
adders | +0.011 |
By | +0.010 |
neys | +0.010 |
online | +0.010 |
Neg logit contributions | |
maid | -0.011 |
fish | -0.011 |
bl | -0.010 |
Qu | -0.010 |
breath | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | +3.13e-01 |
Pos logit contributions | |
supposed | +0.106 |
adders | +0.101 |
By | +0.094 |
neys | +0.094 |
emergencies | +0.091 |
Neg logit contributions | |
fish | -0.111 |
maid | -0.103 |
bl | -0.099 |
breath | -0.094 |
ora | -0.089 |
Token | trees |
Token ID | 7150 |
Feature activation | +5.14e+00 |
Pos logit contributions | |
supposed | +0.014 |
adders | +0.013 |
By | +0.012 |
neys | +0.012 |
online | +0.011 |
Neg logit contributions | |
maid | -0.012 |
fish | -0.011 |
bl | -0.011 |
Qu | -0.010 |
breath | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +3.45e-01 |
Pos logit contributions | |
supposed | +0.232 |
adders | +0.217 |
neys | +0.199 |
By | +0.199 |
emergencies | +0.197 |
Neg logit contributions | |
fish | -0.190 |
bl | -0.170 |
maid | -0.168 |
breath | -0.162 |
Qu | -0.152 |
Token | He |
Token ID | 679 |
Feature activation | -6.62e-01 |
Pos logit contributions | |
supposed | +0.014 |
adders | +0.013 |
done | +0.012 |
measuring | +0.012 |
online | +0.012 |
Neg logit contributions | |
maid | -0.014 |
Qu | -0.013 |
fish | -0.013 |
ora | -0.013 |
dolphin | -0.012 |
Token | felt |
Token ID | 2936 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
maid | +0.029 |
fish | +0.026 |
ora | +0.025 |
Qu | +0.025 |
breath | +0.024 |
Neg logit contributions | |
adders | -0.024 |
supposed | -0.024 |
neys | -0.021 |
about | -0.021 |
done | -0.021 |
Token | disappointed |
Token ID | 11679 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
maid | +0.060 |
fish | +0.059 |
bl | +0.057 |
Qu | +0.057 |
breath | +0.056 |
Neg logit contributions | |
supposed | -0.046 |
adders | -0.044 |
neys | -0.040 |
By | -0.038 |
online | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
supposed | +0.005 |
adders | +0.005 |
neys | +0.004 |
online | +0.004 |
emergencies | +0.004 |
Neg logit contributions | |
maid | -0.004 |
fish | -0.004 |
ora | -0.004 |
bl | -0.004 |
breath | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | -8.16e-01 |
Pos logit contributions | |
bl | +0.046 |
fish | +0.044 |
maid | +0.044 |
dolphin | +0.044 |
breath | +0.044 |
Neg logit contributions | |
adders | -0.031 |
supposed | -0.028 |
neys | -0.025 |
By | -0.023 |
Until | -0.022 |
Token | play |
Token ID | 711 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
maid | +0.032 |
fish | +0.030 |
Qu | +0.029 |
ora | +0.028 |
bl | +0.027 |
Neg logit contributions | |
supposed | -0.034 |
adders | -0.034 |
neys | -0.029 |
about | -0.029 |
online | -0.029 |
Token | with |
Token ID | 351 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
maid | +0.127 |
ora | +0.118 |
fish | +0.116 |
Qu | +0.112 |
breath | +0.109 |
Neg logit contributions | |
adders | -0.041 |
supposed | -0.038 |
about | -0.036 |
driving | -0.034 |
online | -0.034 |
Token | his |
Token ID | 465 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
maid | +0.095 |
fish | +0.090 |
Qu | +0.086 |
bl | +0.085 |
ora | +0.084 |
Neg logit contributions | |
supposed | -0.079 |
adders | -0.078 |
neys | -0.068 |
By | -0.065 |
about | -0.065 |
Token | toy |
Token ID | 13373 |
Feature activation | -9.57e-01 |
Pos logit contributions | |
maid | +0.060 |
fish | +0.057 |
bl | +0.054 |
Qu | +0.053 |
breath | +0.053 |
Neg logit contributions | |
supposed | -0.049 |
adders | -0.049 |
By | -0.042 |
neys | -0.042 |
about | -0.041 |
Token | cars |
Token ID | 5006 |
Feature activation | +5.49e+00 |
Pos logit contributions | |
maid | +0.052 |
fish | +0.049 |
Qu | +0.048 |
ora | +0.048 |
bl | +0.046 |
Neg logit contributions | |
supposed | -0.025 |
adders | -0.025 |
neys | -0.020 |
online | -0.020 |
about | -0.019 |
Token | all |
Token ID | 477 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
supposed | +0.240 |
adders | +0.224 |
emergencies | +0.204 |
By | +0.201 |
neys | +0.199 |
Neg logit contributions | |
fish | -0.203 |
bl | -0.191 |
maid | -0.188 |
Qu | -0.179 |
ora | -0.174 |
Token | day |
Token ID | 1110 |
Feature activation | -7.85e-01 |
Pos logit contributions | |
fish | +0.061 |
maid | +0.060 |
ora | +0.057 |
Qu | +0.057 |
breath | +0.056 |
Neg logit contributions | |
adders | -0.040 |
supposed | -0.036 |
neys | -0.031 |
emergencies | -0.031 |
By | -0.030 |
Token | long |
Token ID | 890 |
Feature activation | +2.77e-01 |
Pos logit contributions | |
maid | +0.038 |
fish | +0.037 |
bl | +0.035 |
Qu | +0.035 |
ora | +0.035 |
Neg logit contributions | |
adders | -0.024 |
supposed | -0.024 |
neys | -0.021 |
online | -0.019 |
By | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -7.15e-02 |
Pos logit contributions | |
supposed | +0.010 |
adders | +0.010 |
neys | +0.009 |
measuring | +0.008 |
online | +0.008 |
Neg logit contributions | |
maid | -0.012 |
fish | -0.012 |
bl | -0.011 |
Qu | -0.011 |
breath | -0.011 |
Token | One |
Token ID | 1881 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
Qu | +0.002 |
maid | +0.002 |
bl | +0.002 |
ora | +0.002 |
fish | +0.002 |
Neg logit contributions | |
supposed | -0.003 |
adders | -0.003 |
done | -0.003 |
eps | -0.003 |
neys | -0.003 |
Token | their |
Token ID | 511 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
maid | +0.025 |
fish | +0.024 |
Qu | +0.023 |
ora | +0.023 |
bl | +0.022 |
Neg logit contributions | |
supposed | -0.015 |
adders | -0.015 |
neys | -0.013 |
about | -0.012 |
online | -0.012 |
Token | parents |
Token ID | 3397 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
maid | +0.010 |
fish | +0.009 |
breath | +0.009 |
bl | +0.009 |
ora | +0.009 |
Neg logit contributions | |
supposed | -0.009 |
adders | -0.009 |
By | -0.007 |
about | -0.007 |
neys | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
adders | +0.009 |
supposed | +0.009 |
neys | +0.009 |
emergencies | +0.008 |
eps | +0.008 |
Neg logit contributions | |
maid | -0.007 |
fish | -0.007 |
bl | -0.006 |
breath | -0.006 |
ora | -0.006 |
Token | told |
Token ID | 1297 |
Feature activation | +7.80e-01 |
Pos logit contributions | |
maid | +0.005 |
fish | +0.005 |
bl | +0.005 |
breath | +0.005 |
ora | +0.005 |
Neg logit contributions | |
supposed | -0.006 |
adders | -0.006 |
neys | -0.005 |
By | -0.005 |
emergencies | -0.005 |
Token | them |
Token ID | 606 |
Feature activation | +9.45e-01 |
Pos logit contributions | |
supposed | +0.034 |
adders | +0.033 |
neys | +0.030 |
By | +0.028 |
online | +0.028 |
Neg logit contributions | |
maid | -0.030 |
fish | -0.028 |
Qu | -0.026 |
bl | -0.026 |
ora | -0.026 |
Token | what |
Token ID | 644 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
supposed | +0.027 |
adders | +0.026 |
neys | +0.023 |
emergencies | +0.021 |
clock | +0.021 |
Neg logit contributions | |
maid | -0.048 |
fish | -0.046 |
bl | -0.044 |
dolphin | -0.043 |
ora | -0.042 |
Token | they |
Token ID | 484 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
adders | +0.079 |
supposed | +0.072 |
neys | +0.066 |
online | +0.065 |
By | +0.063 |
Neg logit contributions | |
fish | -0.092 |
dolphin | -0.086 |
maid | -0.085 |
Qu | -0.083 |
bl | -0.082 |
Token | had |
Token ID | 550 |
Feature activation | +9.22e-01 |
Pos logit contributions | |
adders | +0.001 |
supposed | +0.001 |
neys | +0.001 |
By | +0.001 |
online | +0.001 |
Neg logit contributions | |
maid | -0.001 |
fish | -0.001 |
Qu | -0.001 |
ora | -0.001 |
bl | -0.001 |
Token | done |
Token ID | 1760 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
adders | +0.020 |
supposed | +0.018 |
neys | +0.016 |
By | +0.015 |
chairs | +0.014 |
Neg logit contributions | |
maid | -0.052 |
fish | -0.051 |
bl | -0.049 |
ora | -0.049 |
Qu | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | -2.98e-01 |
Pos logit contributions | |
adders | +0.053 |
supposed | +0.048 |
neys | +0.045 |
Wil | +0.045 |
By | +0.044 |
Neg logit contributions | |
fish | -0.039 |
maid | -0.038 |
bl | -0.034 |
ora | -0.033 |
breath | -0.031 |
Token | Their |
Token ID | 5334 |
Feature activation | -4.37e-02 |
Pos logit contributions | |
maid | +0.012 |
Qu | +0.011 |
ora | +0.011 |
dolphin | +0.011 |
fish | +0.010 |
Neg logit contributions | |
supposed | -0.013 |
adders | -0.012 |
done | -0.010 |
measuring | -0.010 |
neys | -0.010 |
Token | ! |
Token ID | 0 |
Feature activation | +3.95e-01 |
Pos logit contributions | |
maid | +0.020 |
fish | +0.020 |
ora | +0.019 |
bl | +0.018 |
Qu | +0.018 |
Neg logit contributions | |
supposed | -0.013 |
adders | -0.013 |
neys | -0.011 |
online | -0.011 |
By | -0.011 |
Token | I |
Token ID | 314 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
supposed | +0.017 |
adders | +0.017 |
done | +0.015 |
neys | +0.014 |
online | +0.014 |
Neg logit contributions | |
maid | -0.015 |
Qu | -0.014 |
fish | -0.013 |
ora | -0.013 |
bl | -0.013 |
Token | want |
Token ID | 765 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
maid | +0.045 |
fish | +0.042 |
Qu | +0.040 |
ora | +0.039 |
bl | +0.039 |
Neg logit contributions | |
supposed | -0.038 |
adders | -0.036 |
about | -0.033 |
done | -0.032 |
neys | -0.031 |
Token | to |
Token ID | 284 |
Feature activation | -4.53e-01 |
Pos logit contributions | |
adders | +0.046 |
supposed | +0.044 |
neys | +0.039 |
By | +0.035 |
eps | +0.035 |
Neg logit contributions | |
maid | -0.063 |
fish | -0.062 |
bl | -0.060 |
dolphin | -0.060 |
Qu | -0.057 |
Token | go |
Token ID | 467 |
Feature activation | +6.05e-01 |
Pos logit contributions | |
maid | +0.020 |
fish | +0.019 |
Qu | +0.018 |
ora | +0.018 |
bl | +0.017 |
Neg logit contributions | |
supposed | -0.016 |
adders | -0.016 |
about | -0.014 |
measuring | -0.014 |
done | -0.014 |
Token | see |
Token ID | 766 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
adders | +0.019 |
supposed | +0.019 |
neys | +0.016 |
By | +0.015 |
emergencies | +0.015 |
Neg logit contributions | |
maid | -0.030 |
fish | -0.029 |
ora | -0.028 |
bl | -0.027 |
Qu | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
supposed | +0.069 |
adders | +0.067 |
neys | +0.062 |
By | +0.058 |
done | +0.057 |
Neg logit contributions | |
maid | -0.061 |
fish | -0.058 |
bl | -0.055 |
dolphin | -0.054 |
Qu | -0.052 |
Token | land |
Token ID | 1956 |
Feature activation | -8.78e-01 |
Pos logit contributions | |
supposed | +0.042 |
adders | +0.041 |
neys | +0.036 |
about | +0.035 |
online | +0.035 |
Neg logit contributions | |
maid | -0.043 |
fish | -0.041 |
Qu | -0.039 |
ora | -0.039 |
bl | -0.038 |
Token | !" |
Token ID | 2474 |
Feature activation | +5.95e-02 |
Pos logit contributions | |
fish | +0.037 |
maid | +0.036 |
ora | +0.032 |
bl | +0.031 |
breath | +0.031 |
Neg logit contributions | |
supposed | -0.034 |
adders | -0.034 |
emergencies | -0.030 |
neys | -0.029 |
By | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | +6.49e-01 |
Pos logit contributions | |
supposed | +0.002 |
adders | +0.002 |
neys | +0.001 |
about | +0.001 |
online | +0.001 |
Neg logit contributions | |
maid | -0.003 |
fish | -0.003 |
Qu | -0.003 |
ora | -0.003 |
bl | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +6.88e-01 |
Pos logit contributions | |
supposed | +0.018 |
adders | +0.018 |
neys | +0.015 |
about | +0.014 |
online | +0.014 |
Neg logit contributions | |
maid | -0.034 |
fish | -0.032 |
Qu | -0.031 |
ora | -0.031 |
bl | -0.030 |
Token | ila |
Token ID | 10102 |
Feature activation | +3.68e-01 |
Pos logit contributions | |
adders | +0.056 |
supposed | +0.056 |
about | +0.049 |
done | +0.048 |
neys | +0.046 |
Neg logit contributions | |
maid | -0.058 |
fish | -0.054 |
breath | -0.052 |
Qu | -0.052 |
bl | -0.051 |
Token | felt |
Token ID | 2936 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
supposed | +0.014 |
adders | +0.014 |
about | +0.012 |
neys | +0.012 |
done | +0.012 |
Neg logit contributions | |
maid | -0.015 |
fish | -0.014 |
Qu | -0.014 |
ora | -0.014 |
breath | -0.013 |
Token | sad |
Token ID | 6507 |
Feature activation | -9.47e-01 |
Pos logit contributions | |
maid | +0.083 |
fish | +0.079 |
ora | +0.076 |
Qu | +0.075 |
bl | +0.073 |
Neg logit contributions | |
supposed | -0.050 |
adders | -0.048 |
about | -0.040 |
neys | -0.040 |
online | -0.039 |
Token | because |
Token ID | 780 |
Feature activation | +8.49e-01 |
Pos logit contributions | |
maid | +0.048 |
fish | +0.046 |
Qu | +0.044 |
ora | +0.044 |
bl | +0.042 |
Neg logit contributions | |
supposed | -0.029 |
adders | -0.028 |
about | -0.025 |
online | -0.023 |
By | -0.023 |
Token | she |
Token ID | 673 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
supposed | +0.033 |
adders | +0.032 |
neys | +0.029 |
online | +0.027 |
By | +0.026 |
Neg logit contributions | |
maid | -0.034 |
fish | -0.034 |
bl | -0.032 |
Qu | -0.032 |
ora | -0.031 |
Token | wanted |
Token ID | 2227 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
supposed | +0.007 |
adders | +0.006 |
about | +0.005 |
done | +0.005 |
neys | +0.005 |
Neg logit contributions | |
maid | -0.008 |
fish | -0.008 |
ora | -0.007 |
Qu | -0.007 |
breath | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | -8.63e-01 |
Pos logit contributions | |
adders | +0.041 |
supposed | +0.037 |
neys | +0.034 |
By | +0.032 |
online | +0.030 |
Neg logit contributions | |
maid | -0.057 |
bl | -0.057 |
fish | -0.055 |
Qu | -0.054 |
dolphin | -0.053 |
Token | fancy |
Token ID | 14996 |
Feature activation | +6.60e-01 |
Pos logit contributions | |
maid | +0.043 |
ora | +0.042 |
fish | +0.041 |
Qu | +0.040 |
breath | +0.038 |
Neg logit contributions | |
supposed | -0.023 |
measuring | -0.020 |
about | -0.019 |
online | -0.019 |
ready | -0.017 |
Token | toy |
Token ID | 13373 |
Feature activation | -8.64e-01 |
Pos logit contributions | |
supposed | +0.020 |
measuring | +0.020 |
about | +0.019 |
online | +0.019 |
fixing | +0.018 |
Neg logit contributions | |
maid | -0.028 |
Qu | -0.028 |
ora | -0.027 |
fish | -0.026 |
of | -0.025 |
Token | too |
Token ID | 1165 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
maid | +0.043 |
fish | +0.041 |
Qu | +0.039 |
ora | +0.039 |
bl | +0.038 |
Neg logit contributions | |
supposed | -0.027 |
adders | -0.027 |
neys | -0.022 |
By | -0.021 |
about | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -3.36e-01 |
Pos logit contributions | |
maid | +0.008 |
fish | +0.008 |
bl | +0.007 |
ora | +0.007 |
breath | +0.007 |
Neg logit contributions | |
adders | -0.008 |
supposed | -0.007 |
neys | -0.007 |
emergencies | -0.006 |
By | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -6.50e-01 |
Pos logit contributions | |
fish | +0.052 |
maid | +0.051 |
bl | +0.049 |
ora | +0.047 |
lizard | +0.045 |
Neg logit contributions | |
supposed | -0.057 |
adders | -0.056 |
Wil | -0.051 |
emergencies | -0.050 |
measuring | -0.049 |
Token | She |
Token ID | 1375 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
maid | +0.024 |
Qu | +0.022 |
ora | +0.022 |
bl | +0.022 |
dolphin | +0.021 |
Neg logit contributions | |
supposed | -0.028 |
adders | -0.027 |
done | -0.024 |
neys | -0.024 |
measuring | -0.023 |
Token | says |
Token ID | 1139 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
maid | +0.048 |
fish | +0.046 |
bl | +0.044 |
ora | +0.043 |
Qu | +0.043 |
Neg logit contributions | |
supposed | -0.050 |
adders | -0.049 |
neys | -0.044 |
measuring | -0.042 |
By | -0.041 |
Token | she |
Token ID | 673 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
supposed | +0.072 |
adders | +0.066 |
neys | +0.063 |
eps | +0.055 |
done | +0.055 |
Neg logit contributions | |
maid | -0.077 |
fish | -0.074 |
bl | -0.073 |
dolphin | -0.070 |
Qu | -0.070 |
Token | likes |
Token ID | 7832 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
maid | +0.006 |
fish | +0.005 |
bl | +0.005 |
ora | +0.005 |
Qu | +0.005 |
Neg logit contributions | |
adders | -0.007 |
supposed | -0.007 |
neys | -0.007 |
measuring | -0.007 |
By | -0.007 |
Token | them |
Token ID | 606 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
maid | +0.062 |
fish | +0.060 |
bl | +0.059 |
dolphin | +0.057 |
Qu | +0.057 |
Neg logit contributions | |
adders | -0.057 |
supposed | -0.056 |
neys | -0.049 |
By | -0.047 |
done | -0.045 |
Token | very |
Token ID | 845 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
adders | +0.045 |
supposed | +0.045 |
neys | +0.039 |
By | +0.038 |
emergencies | +0.037 |
Neg logit contributions | |
maid | -0.041 |
bl | -0.040 |
fish | -0.038 |
lizard | -0.035 |
ora | -0.035 |
Token | much |
Token ID | 881 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
supposed | +0.003 |
adders | +0.003 |
neys | +0.002 |
measuring | +0.002 |
By | +0.002 |
Neg logit contributions | |
maid | -0.013 |
fish | -0.012 |
ora | -0.012 |
Qu | -0.012 |
bl | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -6.99e-01 |
Pos logit contributions | |
supposed | +0.097 |
adders | +0.095 |
neys | +0.082 |
measuring | +0.081 |
Wil | +0.080 |
Neg logit contributions | |
maid | -0.094 |
fish | -0.092 |
bl | -0.088 |
ora | -0.084 |
lizard | -0.082 |
Token |
|
Token ID | 198 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
maid | +0.026 |
Qu | +0.025 |
ora | +0.024 |
bl | +0.024 |
fish | +0.023 |
Neg logit contributions | |
supposed | -0.030 |
adders | -0.028 |
neys | -0.025 |
done | -0.025 |
measuring | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | -3.48e-01 |
Pos logit contributions | |
maid | +0.020 |
fish | +0.020 |
Qu | +0.019 |
ora | +0.019 |
bl | +0.018 |
Neg logit contributions | |
supposed | -0.014 |
adders | -0.013 |
neys | -0.011 |
measuring | -0.011 |
online | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
maid | +0.008 |
fish | +0.007 |
Qu | +0.007 |
bl | +0.007 |
ora | +0.007 |
Neg logit contributions | |
supposed | -0.008 |
adders | -0.008 |
neys | -0.007 |
online | -0.006 |
measuring | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +2.36e-01 |
Pos logit contributions | |
supposed | +0.005 |
adders | +0.005 |
neys | +0.004 |
about | +0.004 |
online | +0.004 |
Neg logit contributions | |
maid | -0.011 |
fish | -0.010 |
Qu | -0.010 |
ora | -0.010 |
bl | -0.010 |
Token | L |
Token ID | 43 |
Feature activation | -8.93e-01 |
Pos logit contributions | |
supposed | +0.010 |
adders | +0.010 |
By | +0.010 |
neys | +0.009 |
about | +0.009 |
Neg logit contributions | |
maid | -0.009 |
fish | -0.008 |
breath | -0.007 |
ora | -0.007 |
dolphin | -0.007 |
Token | ily |
Token ID | 813 |
Feature activation | -7.54e-01 |
Pos logit contributions | |
maid | +0.052 |
fish | +0.049 |
Qu | +0.048 |
breath | +0.048 |
dolphin | +0.047 |
Neg logit contributions | |
adders | -0.022 |
supposed | -0.022 |
about | -0.019 |
done | -0.016 |
online | -0.016 |
Token | loved |
Token ID | 6151 |
Feature activation | -7.27e-01 |
Pos logit contributions | |
maid | +0.032 |
fish | +0.030 |
Qu | +0.029 |
ora | +0.029 |
breath | +0.028 |
Neg logit contributions | |
supposed | -0.026 |
adders | -0.026 |
about | -0.025 |
done | -0.023 |
neys | -0.023 |
Token | puzzles |
Token ID | 24367 |
Feature activation | +3.13e+00 |
Pos logit contributions | |
maid | +0.032 |
fish | +0.030 |
bl | +0.030 |
breath | +0.029 |
Qu | +0.029 |
Neg logit contributions | |
adders | -0.027 |
supposed | -0.026 |
neys | -0.023 |
By | -0.021 |
done | -0.021 |
Token | , |
Token ID | 11 |
Feature activation | -2.81e-01 |
Pos logit contributions | |
supposed | +0.112 |
adders | +0.111 |
neys | +0.096 |
By | +0.091 |
emergencies | +0.091 |
Neg logit contributions | |
maid | -0.139 |
fish | -0.134 |
bl | -0.126 |
ora | -0.125 |
Qu | -0.125 |
Token | so |
Token ID | 523 |
Feature activation | +3.06e-01 |
Pos logit contributions | |
maid | +0.012 |
bl | +0.011 |
fish | +0.011 |
Qu | +0.011 |
breath | +0.011 |
Neg logit contributions | |
supposed | -0.010 |
adders | -0.010 |
neys | -0.009 |
online | -0.008 |
emergencies | -0.008 |
Token | she |
Token ID | 673 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
supposed | +0.012 |
adders | +0.011 |
neys | +0.010 |
online | +0.009 |
By | +0.009 |
Neg logit contributions | |
maid | -0.013 |
fish | -0.013 |
bl | -0.012 |
Qu | -0.012 |
ora | -0.012 |
Token | said |
Token ID | 531 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
supposed | +0.008 |
adders | +0.008 |
about | +0.007 |
online | +0.007 |
neys | +0.007 |
Neg logit contributions | |
maid | -0.010 |
fish | -0.010 |
ora | -0.009 |
Qu | -0.009 |
breath | -0.009 |
Token | yes |
Token ID | 3763 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
supposed | +0.072 |
adders | +0.072 |
about | +0.059 |
neys | +0.059 |
online | +0.058 |
Neg logit contributions | |
maid | -0.116 |
fish | -0.110 |
Qu | -0.106 |
ora | -0.105 |
bl | -0.102 |
Token | Once |
Token ID | 7454 |
Feature activation | -2.77e+00 |
Pos logit contributions | |
maid | +0.150 |
fish | +0.145 |
Qu | +0.142 |
ora | +0.142 |
bl | +0.135 |
Neg logit contributions | |
supposed | -0.115 |
adders | -0.113 |
neys | -0.096 |
online | -0.094 |
measuring | -0.091 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
maid | +0.104 |
fish | +0.099 |
ora | +0.093 |
Qu | +0.088 |
breath | +0.088 |
Neg logit contributions | |
supposed | -0.121 |
adders | -0.119 |
neys | -0.107 |
emergencies | -0.102 |
online | -0.102 |
Token | a |
Token ID | 257 |
Feature activation | -7.66e-01 |
Pos logit contributions | |
maid | +0.066 |
fish | +0.064 |
Qu | +0.062 |
ora | +0.061 |
bl | +0.060 |
Neg logit contributions | |
supposed | -0.027 |
adders | -0.027 |
neys | -0.021 |
online | -0.020 |
By | -0.020 |
Token | time |
Token ID | 640 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
maid | +0.043 |
fish | +0.042 |
Qu | +0.040 |
ora | +0.040 |
bl | +0.039 |
Neg logit contributions | |
supposed | -0.018 |
adders | -0.018 |
neys | -0.014 |
online | -0.013 |
about | -0.013 |
Token | there |
Token ID | 612 |
Feature activation | -6.47e-02 |
Pos logit contributions | |
maid | +0.005 |
fish | +0.005 |
ora | +0.004 |
Qu | +0.004 |
bl | +0.004 |
Neg logit contributions | |
supposed | -0.005 |
adders | -0.004 |
neys | -0.004 |
online | -0.004 |
emergencies | -0.004 |
Token | was |
Token ID | 373 |
Feature activation | +6.60e-01 |
Pos logit contributions | |
maid | +0.003 |
fish | +0.002 |
Qu | +0.002 |
ora | +0.002 |
bl | +0.002 |
Neg logit contributions | |
supposed | -0.003 |
adders | -0.003 |
neys | -0.002 |
about | -0.002 |
online | -0.002 |
Token | an |
Token ID | 281 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
adders | +0.023 |
neys | +0.023 |
supposed | +0.022 |
cases | +0.020 |
chairs | +0.019 |
Neg logit contributions | |
Qu | -0.028 |
dolphin | -0.027 |
maid | -0.027 |
bl | -0.026 |
fish | -0.026 |
Token | aer |
Token ID | 9551 |
Feature activation | +3.54e+00 |
Pos logit contributions | |
maid | +0.051 |
fish | +0.049 |
bl | +0.048 |
dolphin | +0.048 |
ora | +0.047 |
Neg logit contributions | |
adders | -0.038 |
supposed | -0.037 |
neys | -0.033 |
By | -0.032 |
done | -0.031 |
Token | opl |
Token ID | 20106 |
Feature activation | +6.71e+00 |
Pos logit contributions | |
supposed | +0.059 |
adders | +0.058 |
neys | +0.040 |
about | +0.037 |
online | +0.037 |
Neg logit contributions | |
maid | -0.227 |
fish | -0.218 |
Qu | -0.211 |
ora | -0.210 |
bl | -0.207 |
Token | ane |
Token ID | 1531 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
supposed | +0.220 |
adders | +0.216 |
neys | +0.183 |
about | +0.178 |
online | +0.178 |
Neg logit contributions | |
maid | -0.322 |
fish | -0.306 |
Qu | -0.293 |
ora | -0.291 |
bl | -0.285 |
Token | that |
Token ID | 326 |
Feature activation | -9.34e-01 |
Pos logit contributions | |
supposed | +0.059 |
adders | +0.057 |
neys | +0.048 |
about | +0.047 |
online | +0.046 |
Neg logit contributions | |
maid | -0.102 |
fish | -0.096 |
Qu | -0.093 |
ora | -0.093 |
bl | -0.090 |
Token | have |
Token ID | 423 |
Feature activation | +8.79e-01 |
Pos logit contributions | |
adders | +0.017 |
supposed | +0.017 |
neys | +0.015 |
By | +0.015 |
emergencies | +0.014 |
Neg logit contributions | |
maid | -0.011 |
bl | -0.011 |
breath | -0.010 |
fish | -0.010 |
Qu | -0.010 |
Token | breakfast |
Token ID | 12607 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
adders | +0.032 |
supposed | +0.030 |
neys | +0.029 |
penter | +0.026 |
By | +0.026 |
Neg logit contributions | |
bl | -0.036 |
dolphin | -0.033 |
jelly | -0.033 |
breath | -0.033 |
maid | -0.033 |
Token | together |
Token ID | 1978 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
fish | +0.001 |
maid | +0.001 |
bl | +0.001 |
breath | +0.001 |
ora | +0.001 |
Neg logit contributions | |
adders | -0.001 |
supposed | -0.001 |
neys | -0.001 |
emergencies | -0.001 |
elev | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
adders | +0.043 |
supposed | +0.043 |
neys | +0.038 |
emergencies | +0.037 |
By | +0.036 |
Neg logit contributions | |
fish | -0.044 |
maid | -0.044 |
bl | -0.041 |
breath | -0.040 |
ora | -0.039 |
Token | |
Token ID | 220 |
Feature activation | -5.22e-01 |
Pos logit contributions | |
supposed | +0.004 |
adders | +0.004 |
fixing | +0.004 |
emergencies | +0.004 |
driving | +0.004 |
Neg logit contributions | |
ora | -0.004 |
maid | -0.004 |
Qu | -0.004 |
of | -0.004 |
lizard | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +6.72e-01 |
Pos logit contributions | |
maid | +0.016 |
ora | +0.016 |
Qu | +0.015 |
fish | +0.015 |
of | +0.014 |
Neg logit contributions | |
supposed | -0.023 |
adders | -0.021 |
builders | -0.020 |
emergencies | -0.020 |
driving | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | +6.96e-01 |
Pos logit contributions | |
supposed | +0.022 |
adders | +0.021 |
neys | +0.018 |
online | +0.017 |
emergencies | +0.017 |
Neg logit contributions | |
maid | -0.031 |
fish | -0.031 |
ora | -0.030 |
Qu | -0.030 |
breath | -0.028 |
Token | After |
Token ID | 3260 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
adders | +0.031 |
supposed | +0.031 |
fixing | +0.028 |
emergencies | +0.027 |
neys | +0.027 |
Neg logit contributions | |
Qu | -0.022 |
ora | -0.022 |
fish | -0.021 |
maid | -0.021 |
of | -0.019 |
Token | breakfast |
Token ID | 12607 |
Feature activation | -5.41e-02 |
Pos logit contributions | |
adders | +0.102 |
supposed | +0.099 |
neys | +0.096 |
eps | +0.081 |
online | +0.079 |
Neg logit contributions | |
bl | -0.107 |
maid | -0.105 |
fish | -0.104 |
dolphin | -0.103 |
breath | -0.102 |
Token | , |
Token ID | 11 |
Feature activation | -8.83e-02 |
Pos logit contributions | |
maid | +0.002 |
fish | +0.002 |
bl | +0.002 |
Qu | +0.002 |
ora | +0.002 |
Neg logit contributions | |
supposed | -0.002 |
adders | -0.002 |
neys | -0.002 |
online | -0.002 |
emergencies | -0.002 |
Token | Mary |
Token ID | 5335 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
maid | +0.004 |
fish | +0.004 |
ora | +0.004 |
breath | +0.004 |
dolphin | +0.004 |
Neg logit contributions | |
supposed | -0.003 |
adders | -0.003 |
neys | -0.002 |
emergencies | -0.002 |
eps | -0.002 |
Token | Tom |
Token ID | 13787 |
Feature activation | +5.93e-01 |
Pos logit contributions | |
maid | +0.007 |
fish | +0.007 |
Qu | +0.007 |
ora | +0.006 |
bl | +0.006 |
Neg logit contributions | |
supposed | -0.006 |
adders | -0.006 |
neys | -0.005 |
about | -0.005 |
online | -0.005 |
Token | was |
Token ID | 373 |
Feature activation | +3.77e-01 |
Pos logit contributions | |
supposed | +0.018 |
about | +0.018 |
adders | +0.018 |
done | +0.017 |
counted | +0.017 |
Neg logit contributions | |
maid | -0.029 |
fish | -0.027 |
ora | -0.026 |
Qu | -0.025 |
dolphin | -0.025 |
Token | surprised |
Token ID | 6655 |
Feature activation | -4.76e-01 |
Pos logit contributions | |
adders | +0.013 |
supposed | +0.012 |
neys | +0.012 |
online | +0.011 |
By | +0.011 |
Neg logit contributions | |
maid | -0.017 |
bl | -0.016 |
Qu | -0.015 |
fish | -0.015 |
breath | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
maid | +0.026 |
fish | +0.025 |
Qu | +0.024 |
ora | +0.024 |
bl | +0.023 |
Neg logit contributions | |
supposed | -0.012 |
adders | -0.011 |
about | -0.011 |
neys | -0.010 |
done | -0.009 |
Token | He |
Token ID | 679 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
maid | +0.006 |
Qu | +0.006 |
fish | +0.006 |
ora | +0.006 |
bl | +0.006 |
Neg logit contributions | |
supposed | -0.005 |
adders | -0.005 |
neys | -0.004 |
done | -0.004 |
measuring | -0.004 |
Token | did |
Token ID | 750 |
Feature activation | -6.68e-01 |
Pos logit contributions | |
maid | +0.061 |
fish | +0.057 |
ora | +0.057 |
breath | +0.052 |
dolphin | +0.051 |
Neg logit contributions | |
supposed | -0.034 |
about | -0.033 |
done | -0.031 |
counted | -0.031 |
adders | -0.030 |
Token | not |
Token ID | 407 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
maid | +0.036 |
fish | +0.034 |
Qu | +0.033 |
ora | +0.032 |
bl | +0.032 |
Neg logit contributions | |
supposed | -0.018 |
adders | -0.018 |
neys | -0.015 |
By | -0.014 |
online | -0.014 |
Token | think |
Token ID | 892 |
Feature activation | +8.21e-01 |
Pos logit contributions | |
adders | +0.005 |
supposed | +0.005 |
neys | +0.004 |
By | +0.004 |
online | +0.004 |
Neg logit contributions | |
maid | -0.006 |
fish | -0.006 |
Qu | -0.006 |
ora | -0.006 |
bl | -0.006 |
Token | Lily |
Token ID | 20037 |
Feature activation | -5.08e-01 |
Pos logit contributions | |
supposed | +0.027 |
neys | +0.026 |
adders | +0.024 |
Wil | +0.023 |
clock | +0.022 |
Neg logit contributions | |
fish | -0.037 |
bl | -0.033 |
maid | -0.033 |
dolphin | -0.032 |
lizard | -0.031 |
Token | would |
Token ID | 561 |
Feature activation | -6.44e-01 |
Pos logit contributions | |
maid | +0.020 |
fish | +0.019 |
Qu | +0.017 |
bl | +0.017 |
ora | +0.017 |
Neg logit contributions | |
adders | -0.021 |
supposed | -0.021 |
neys | -0.018 |
By | -0.018 |
online | -0.018 |
Token | mind |
Token ID | 2000 |
Feature activation | -2.72e+00 |
Pos logit contributions | |
maid | +0.029 |
fish | +0.027 |
ora | +0.026 |
Qu | +0.026 |
bl | +0.025 |
Neg logit contributions | |
supposed | -0.024 |
adders | -0.023 |
neys | -0.020 |
about | -0.020 |
online | -0.019 |
Token | hipp |
Token ID | 18568 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
adders | +0.026 |
supposed | +0.025 |
online | +0.022 |
neys | +0.022 |
By | +0.021 |
Neg logit contributions | |
maid | -0.023 |
fish | -0.023 |
Qu | -0.023 |
bl | -0.022 |
breath | -0.022 |
Token | o |
Token ID | 78 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
adders | +0.069 |
supposed | +0.069 |
emergencies | +0.059 |
online | +0.059 |
By | +0.055 |
Neg logit contributions | |
fish | -0.109 |
maid | -0.108 |
Qu | -0.101 |
ora | -0.100 |
bl | -0.094 |
Token | never |
Token ID | 1239 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
maid | +0.014 |
fish | +0.014 |
Qu | +0.013 |
ora | +0.012 |
bl | +0.012 |
Neg logit contributions | |
adders | -0.015 |
supposed | -0.015 |
neys | -0.013 |
online | -0.012 |
By | -0.012 |
Token | forgot |
Token ID | 16453 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
maid | +0.008 |
fish | +0.007 |
Qu | +0.007 |
ora | +0.007 |
bl | +0.007 |
Neg logit contributions | |
adders | -0.008 |
supposed | -0.008 |
neys | -0.007 |
online | -0.007 |
By | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | +5.89e-01 |
Pos logit contributions | |
maid | +0.009 |
fish | +0.009 |
bl | +0.009 |
dolphin | +0.009 |
snail | +0.008 |
Neg logit contributions | |
adders | -0.009 |
supposed | -0.009 |
neys | -0.008 |
By | -0.008 |
clock | -0.008 |
Token | fast |
Token ID | 3049 |
Feature activation | +3.04e-01 |
Pos logit contributions | |
adders | +0.024 |
supposed | +0.022 |
neys | +0.021 |
online | +0.020 |
By | +0.020 |
Neg logit contributions | |
maid | -0.024 |
fish | -0.023 |
bl | -0.022 |
breath | -0.022 |
Qu | -0.021 |
Token | che |
Token ID | 1125 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
supposed | +0.013 |
adders | +0.013 |
neys | +0.012 |
online | +0.011 |
about | +0.011 |
Neg logit contributions | |
maid | -0.011 |
fish | -0.011 |
Qu | -0.010 |
ora | -0.010 |
bl | -0.010 |
Token | et |
Token ID | 316 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
maid | +0.067 |
Qu | +0.061 |
fish | +0.061 |
ora | +0.059 |
bl | +0.059 |
Neg logit contributions | |
supposed | -0.043 |
adders | -0.042 |
done | -0.036 |
neys | -0.035 |
measuring | -0.034 |
Token | ah |
Token ID | 993 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
adders | +0.101 |
supposed | +0.098 |
emergencies | +0.094 |
By | +0.088 |
firefighters | +0.088 |
Neg logit contributions | |
fish | -0.074 |
maid | -0.069 |
bl | -0.062 |
ora | -0.061 |
lizard | -0.061 |
Token | he |
Token ID | 339 |
Feature activation | -2.29e-01 |
Pos logit contributions | |
fish | +0.014 |
maid | +0.013 |
of | +0.011 |
Qu | +0.011 |
bl | +0.011 |
Neg logit contributions | |
adders | -0.015 |
supposed | -0.014 |
online | -0.014 |
emergencies | -0.013 |
firefighters | -0.013 |
Token | had |
Token ID | 550 |
Feature activation | +9.42e-01 |
Pos logit contributions | |
maid | +0.009 |
fish | +0.009 |
Qu | +0.008 |
ora | +0.008 |
bl | +0.008 |
Neg logit contributions | |
adders | -0.009 |
supposed | -0.009 |
neys | -0.008 |
By | -0.008 |
online | -0.008 |
Token | their |
Token ID | 511 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
adders | +0.011 |
supposed | +0.011 |
neys | +0.010 |
online | +0.009 |
By | +0.009 |
Neg logit contributions | |
maid | -0.015 |
fish | -0.014 |
Qu | -0.014 |
bl | -0.014 |
ora | -0.014 |
Token | poverty |
Token ID | 8098 |
Feature activation | +4.76e-01 |
Pos logit contributions | |
maid | +0.007 |
fish | +0.006 |
bl | +0.006 |
breath | +0.006 |
dolphin | +0.006 |
Neg logit contributions | |
adders | -0.007 |
supposed | -0.007 |
neys | -0.006 |
By | -0.006 |
online | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
adders | +0.018 |
supposed | +0.018 |
neys | +0.015 |
online | +0.014 |
By | +0.014 |
Neg logit contributions | |
maid | -0.021 |
fish | -0.020 |
bl | -0.019 |
Qu | -0.018 |
ora | -0.018 |
Token | all |
Token ID | 477 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
maid | +0.011 |
fish | +0.010 |
bl | +0.010 |
dolphin | +0.010 |
Qu | +0.010 |
Neg logit contributions | |
adders | -0.010 |
supposed | -0.010 |
neys | -0.009 |
By | -0.008 |
online | -0.008 |
Token | of |
Token ID | 286 |
Feature activation | +3.13e-01 |
Pos logit contributions | |
fish | +0.048 |
maid | +0.046 |
bl | +0.044 |
dolphin | +0.043 |
breath | +0.042 |
Neg logit contributions | |
adders | -0.045 |
supposed | -0.043 |
neys | -0.038 |
By | -0.036 |
online | -0.035 |
Token | the |
Token ID | 262 |
Feature activation | +7.12e-01 |
Pos logit contributions | |
adders | +0.012 |
supposed | +0.012 |
neys | +0.010 |
online | +0.010 |
By | +0.010 |
Neg logit contributions | |
maid | -0.013 |
fish | -0.013 |
Qu | -0.012 |
bl | -0.012 |
ora | -0.012 |
Token | family |
Token ID | 1641 |
Feature activation | +7.70e-01 |
Pos logit contributions | |
adders | +0.027 |
supposed | +0.026 |
neys | +0.023 |
By | +0.023 |
about | +0.022 |
Neg logit contributions | |
maid | -0.030 |
fish | -0.029 |
bl | -0.028 |
breath | -0.028 |
dolphin | -0.028 |
Token | 's |
Token ID | 338 |
Feature activation | -9.43e-01 |
Pos logit contributions | |
adders | +0.032 |
supposed | +0.031 |
neys | +0.027 |
emergencies | +0.027 |
online | +0.026 |
Neg logit contributions | |
fish | -0.031 |
maid | -0.030 |
bl | -0.028 |
dolphin | -0.026 |
ora | -0.026 |
Token | possessions |
Token ID | 23309 |
Feature activation | +4.32e-01 |
Pos logit contributions | |
maid | +0.036 |
fish | +0.033 |
bl | +0.033 |
breath | +0.033 |
dolphin | +0.031 |
Neg logit contributions | |
adders | -0.040 |
supposed | -0.038 |
online | -0.034 |
By | -0.034 |
neys | -0.034 |
Token | had |
Token ID | 550 |
Feature activation | +8.54e-01 |
Pos logit contributions | |
adders | +0.016 |
supposed | +0.016 |
neys | +0.014 |
By | +0.013 |
emergencies | +0.013 |
Neg logit contributions | |
maid | -0.018 |
fish | -0.018 |
bl | -0.017 |
breath | -0.016 |
Qu | -0.016 |
Token | been |
Token ID | 587 |
Feature activation | -5.71e-01 |
Pos logit contributions | |
adders | +0.028 |
supposed | +0.025 |
neys | +0.024 |
By | +0.022 |
online | +0.022 |
Neg logit contributions | |
maid | -0.039 |
fish | -0.038 |
bl | -0.038 |
Qu | -0.037 |
breath | -0.036 |
Token | a |
Token ID | 257 |
Feature activation | -9.35e-01 |
Pos logit contributions | |
maid | +0.033 |
fish | +0.031 |
Qu | +0.029 |
ora | +0.029 |
bl | +0.029 |
Neg logit contributions | |
supposed | -0.023 |
adders | -0.022 |
neys | -0.019 |
about | -0.018 |
online | -0.018 |
Token | kind |
Token ID | 1611 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
maid | +0.038 |
fish | +0.036 |
Qu | +0.035 |
bl | +0.035 |
dolphin | +0.034 |
Neg logit contributions | |
adders | -0.039 |
supposed | -0.037 |
neys | -0.033 |
By | -0.031 |
about | -0.030 |
Token | person |
Token ID | 1048 |
Feature activation | -8.84e-01 |
Pos logit contributions | |
maid | +0.077 |
fish | +0.074 |
Qu | +0.070 |
ora | +0.067 |
breath | +0.064 |
Neg logit contributions | |
adders | -0.056 |
supposed | -0.056 |
about | -0.055 |
office | -0.052 |
traffic | -0.051 |
Token | who |
Token ID | 508 |
Feature activation | +7.41e-01 |
Pos logit contributions | |
fish | +0.034 |
maid | +0.034 |
bl | +0.031 |
dolphin | +0.029 |
ora | +0.029 |
Neg logit contributions | |
supposed | -0.037 |
adders | -0.037 |
neys | -0.033 |
By | -0.032 |
emergencies | -0.031 |
Token | could |
Token ID | 714 |
Feature activation | -5.33e-01 |
Pos logit contributions | |
supposed | +0.027 |
adders | +0.027 |
neys | +0.023 |
about | +0.023 |
online | +0.023 |
Neg logit contributions | |
maid | -0.033 |
fish | -0.030 |
Qu | -0.029 |
ora | -0.029 |
bl | -0.028 |
Token | help |
Token ID | 1037 |
Feature activation | -3.62e+00 |
Pos logit contributions | |
maid | +0.024 |
fish | +0.021 |
ora | +0.021 |
Qu | +0.020 |
dolphin | +0.019 |
Neg logit contributions | |
supposed | -0.020 |
adders | -0.019 |
about | -0.017 |
measuring | -0.017 |
driving | -0.016 |
Token | it |
Token ID | 340 |
Feature activation | -5.40e-01 |
Pos logit contributions | |
maid | +0.140 |
fish | +0.134 |
bl | +0.128 |
breath | +0.125 |
Qu | +0.125 |
Neg logit contributions | |
adders | -0.150 |
supposed | -0.146 |
neys | -0.131 |
By | -0.127 |
online | -0.123 |
Token | . |
Token ID | 13 |
Feature activation | +9.25e-02 |
Pos logit contributions | |
maid | +0.020 |
fish | +0.020 |
bl | +0.018 |
breath | +0.017 |
Qu | +0.017 |
Neg logit contributions | |
supposed | -0.023 |
adders | -0.023 |
By | -0.020 |
neys | -0.020 |
emergencies | -0.019 |
Token | |
Token ID | 220 |
Feature activation | -8.65e-01 |
Pos logit contributions | |
supposed | +0.004 |
adders | +0.003 |
done | +0.003 |
neys | +0.003 |
eps | +0.003 |
Neg logit contributions | |
maid | -0.004 |
Qu | -0.004 |
ora | -0.004 |
fish | -0.004 |
bl | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +6.19e-01 |
Pos logit contributions | |
maid | +0.029 |
ora | +0.028 |
Qu | +0.028 |
fish | +0.027 |
of | +0.027 |
Neg logit contributions | |
supposed | -0.037 |
adders | -0.033 |
emergencies | -0.032 |
elev | -0.032 |
online | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | +6.43e-01 |
Pos logit contributions | |
supposed | +0.020 |
adders | +0.020 |
neys | +0.016 |
online | +0.016 |
emergencies | +0.016 |
Neg logit contributions | |
maid | -0.029 |
fish | -0.029 |
Qu | -0.029 |
ora | -0.028 |
bl | -0.026 |
Token | water |
Token ID | 1660 |
Feature activation | -8.48e-01 |
Pos logit contributions | |
maid | +0.028 |
fish | +0.027 |
Qu | +0.025 |
ora | +0.024 |
dolphin | +0.024 |
Neg logit contributions | |
supposed | -0.028 |
adders | -0.026 |
done | -0.025 |
about | -0.024 |
measuring | -0.024 |
Token | her |
Token ID | 607 |
Feature activation | -9.00e-01 |
Pos logit contributions | |
maid | +0.040 |
fish | +0.039 |
ora | +0.036 |
Qu | +0.036 |
bl | +0.036 |
Neg logit contributions | |
supposed | -0.028 |
adders | -0.028 |
neys | -0.024 |
By | -0.023 |
online | -0.022 |
Token | plant |
Token ID | 4618 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
maid | +0.037 |
fish | +0.035 |
bl | +0.033 |
breath | +0.032 |
ora | +0.032 |
Neg logit contributions | |
adders | -0.037 |
supposed | -0.037 |
By | -0.032 |
neys | -0.032 |
online | -0.031 |
Token | again |
Token ID | 757 |
Feature activation | +3.02e-02 |
Pos logit contributions | |
maid | +0.087 |
fish | +0.086 |
bl | +0.080 |
ora | +0.077 |
Qu | +0.076 |
Neg logit contributions | |
supposed | -0.071 |
adders | -0.070 |
By | -0.059 |
neys | -0.059 |
emergencies | -0.057 |
Token | . |
Token ID | 13 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
adders | +0.001 |
supposed | +0.001 |
neys | +0.001 |
online | +0.001 |
By | +0.001 |
Neg logit contributions | |
maid | -0.001 |
fish | -0.001 |
ora | -0.001 |
Qu | -0.001 |
bl | -0.001 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.45e+00 |
Pos logit contributions | |
adders | +0.007 |
supposed | +0.007 |
measuring | +0.006 |
neys | +0.006 |
done | +0.005 |
Neg logit contributions | |
maid | -0.010 |
Qu | -0.009 |
ora | -0.009 |
fish | -0.009 |
bl | -0.009 |
Token | One |
Token ID | 3198 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
maid | +0.155 |
fish | +0.152 |
Qu | +0.149 |
ora | +0.146 |
bl | +0.142 |
Neg logit contributions | |
supposed | -0.118 |
adders | -0.117 |
neys | -0.097 |
online | -0.096 |
measuring | -0.094 |
Token | day |
Token ID | 1110 |
Feature activation | -6.28e-01 |
Pos logit contributions | |
adders | +0.049 |
supposed | +0.047 |
neys | +0.041 |
online | +0.039 |
emergencies | +0.039 |
Neg logit contributions | |
fish | -0.076 |
maid | -0.074 |
ora | -0.071 |
Qu | -0.071 |
breath | -0.070 |
Token | , |
Token ID | 11 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
maid | +0.029 |
fish | +0.028 |
Qu | +0.027 |
ora | +0.027 |
bl | +0.026 |
Neg logit contributions | |
supposed | -0.021 |
adders | -0.021 |
neys | -0.019 |
online | -0.018 |
emergencies | -0.017 |
Token | Tom |
Token ID | 4186 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
maid | +0.015 |
fish | +0.014 |
Qu | +0.014 |
ora | +0.014 |
bl | +0.013 |
Neg logit contributions | |
supposed | -0.006 |
adders | -0.006 |
neys | -0.005 |
about | -0.004 |
online | -0.004 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
adders | +0.058 |
supposed | +0.057 |
neys | +0.050 |
online | +0.049 |
By | +0.048 |
Neg logit contributions | |
maid | -0.065 |
fish | -0.062 |
Qu | -0.059 |
ora | -0.059 |
bl | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | +9.36e-02 |
Pos logit contributions | |
fish | +0.016 |
maid | +0.016 |
bl | +0.015 |
dolphin | +0.014 |
breath | +0.014 |
Neg logit contributions | |
adders | -0.013 |
supposed | -0.013 |
neys | -0.012 |
elev | -0.011 |
emergencies | -0.011 |
Token | Lily |
Token ID | 20037 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
supposed | +0.004 |
adders | +0.004 |
neys | +0.003 |
emergencies | +0.003 |
done | +0.003 |
Neg logit contributions | |
maid | -0.004 |
Qu | -0.004 |
bl | -0.004 |
fish | -0.004 |
ora | -0.004 |
Token | had |
Token ID | 550 |
Feature activation | +9.48e-01 |
Pos logit contributions | |
maid | +0.017 |
ora | +0.015 |
fish | +0.015 |
Qu | +0.015 |
breath | +0.015 |
Neg logit contributions | |
supposed | -0.010 |
about | -0.009 |
done | -0.009 |
adders | -0.008 |
counted | -0.008 |
Token | two |
Token ID | 734 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
adders | +0.034 |
supposed | +0.029 |
neys | +0.028 |
online | +0.026 |
emergencies | +0.026 |
Neg logit contributions | |
bl | -0.042 |
maid | -0.039 |
Qu | -0.038 |
breath | -0.038 |
fish | -0.038 |
Token | big |
Token ID | 1263 |
Feature activation | +5.82e-02 |
Pos logit contributions | |
adders | +0.055 |
supposed | +0.054 |
neys | +0.048 |
online | +0.045 |
By | +0.045 |
Neg logit contributions | |
maid | -0.057 |
fish | -0.054 |
bl | -0.054 |
breath | -0.052 |
dolphin | -0.050 |
Token | help |
Token ID | 1037 |
Feature activation | -3.72e+00 |
Pos logit contributions | |
adders | +0.002 |
supposed | +0.002 |
neys | +0.002 |
By | +0.002 |
online | +0.002 |
Neg logit contributions | |
maid | -0.002 |
bl | -0.002 |
fish | -0.002 |
breath | -0.002 |
dolphin | -0.002 |
Token | ings |
Token ID | 654 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
fish | +0.130 |
maid | +0.128 |
bl | +0.126 |
breath | +0.124 |
dolphin | +0.118 |
Neg logit contributions | |
adders | -0.159 |
supposed | -0.145 |
neys | -0.140 |
elev | -0.130 |
eps | -0.130 |
Token | of |
Token ID | 286 |
Feature activation | +4.16e-01 |
Pos logit contributions | |
fish | +0.061 |
bl | +0.061 |
maid | +0.061 |
Qu | +0.058 |
ora | +0.056 |
Neg logit contributions | |
adders | -0.067 |
supposed | -0.067 |
neys | -0.059 |
emergencies | -0.055 |
By | -0.055 |
Token | the |
Token ID | 262 |
Feature activation | +8.37e-01 |
Pos logit contributions | |
adders | +0.016 |
supposed | +0.016 |
neys | +0.014 |
By | +0.013 |
online | +0.013 |
Neg logit contributions | |
maid | -0.018 |
fish | -0.017 |
bl | -0.016 |
breath | -0.016 |
Qu | -0.016 |
Token | stew |
Token ID | 20798 |
Feature activation | -4.07e-01 |
Pos logit contributions | |
adders | +0.039 |
supposed | +0.035 |
neys | +0.033 |
By | +0.031 |
Kim | +0.030 |
Neg logit contributions | |
maid | -0.031 |
fish | -0.030 |
bl | -0.030 |
dolphin | -0.029 |
breath | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | +3.39e-01 |
Pos logit contributions | |
fish | +0.017 |
maid | +0.017 |
bl | +0.016 |
dolphin | +0.014 |
ora | +0.014 |
Neg logit contributions | |
adders | -0.016 |
supposed | -0.016 |
neys | -0.014 |
elev | -0.014 |
emergencies | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
maid | +0.160 |
fish | +0.153 |
Qu | +0.150 |
ora | +0.148 |
bl | +0.143 |
Neg logit contributions | |
supposed | -0.090 |
adders | -0.088 |
about | -0.075 |
neys | -0.072 |
online | -0.071 |
Token | Teddy |
Token ID | 29345 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
maid | +0.015 |
fish | +0.014 |
ora | +0.014 |
Qu | +0.014 |
breath | +0.013 |
Neg logit contributions | |
supposed | -0.011 |
adders | -0.011 |
about | -0.010 |
online | -0.009 |
By | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +9.28e-02 |
Pos logit contributions | |
maid | +0.065 |
fish | +0.063 |
ora | +0.058 |
bl | +0.058 |
dolphin | +0.056 |
Neg logit contributions | |
adders | -0.057 |
supposed | -0.054 |
neys | -0.049 |
By | -0.048 |
elev | -0.047 |
Token | I |
Token ID | 314 |
Feature activation | -9.88e-01 |
Pos logit contributions | |
supposed | +0.004 |
adders | +0.004 |
done | +0.003 |
online | +0.003 |
neys | +0.003 |
Neg logit contributions | |
maid | -0.004 |
fish | -0.003 |
Qu | -0.003 |
ora | -0.003 |
bl | -0.003 |
Token | 'll |
Token ID | 1183 |
Feature activation | -9.65e-01 |
Pos logit contributions | |
maid | +0.042 |
fish | +0.039 |
Qu | +0.038 |
ora | +0.037 |
bl | +0.036 |
Neg logit contributions | |
supposed | -0.038 |
adders | -0.037 |
neys | -0.033 |
about | -0.032 |
online | -0.032 |
Token | hold |
Token ID | 1745 |
Feature activation | -3.56e+00 |
Pos logit contributions | |
maid | +0.042 |
fish | +0.039 |
Qu | +0.038 |
ora | +0.037 |
bl | +0.036 |
Neg logit contributions | |
supposed | -0.037 |
adders | -0.036 |
neys | -0.031 |
about | -0.031 |
measuring | -0.030 |
Token | your |
Token ID | 534 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
maid | +0.151 |
fish | +0.143 |
breath | +0.140 |
bl | +0.137 |
Qu | +0.137 |
Neg logit contributions | |
supposed | -0.138 |
adders | -0.130 |
neys | -0.120 |
about | -0.114 |
By | -0.112 |
Token | handle |
Token ID | 5412 |
Feature activation | +5.10e-01 |
Pos logit contributions | |
maid | +0.050 |
fish | +0.048 |
ora | +0.048 |
Qu | +0.047 |
dolphin | +0.042 |
Neg logit contributions | |
supposed | -0.055 |
adders | -0.055 |
measuring | -0.049 |
online | -0.048 |
neys | -0.048 |
Token | and |
Token ID | 290 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
supposed | +0.024 |
adders | +0.021 |
emergencies | +0.021 |
By | +0.020 |
discussing | +0.019 |
Neg logit contributions | |
fish | -0.017 |
maid | -0.017 |
breath | -0.016 |
bl | -0.016 |
ora | -0.014 |
Token | we |
Token ID | 356 |
Feature activation | +8.93e-01 |
Pos logit contributions | |
supposed | +0.007 |
adders | +0.007 |
neys | +0.006 |
By | +0.006 |
about | +0.006 |
Neg logit contributions | |
maid | -0.005 |
fish | -0.005 |
breath | -0.005 |
bl | -0.005 |
Qu | -0.004 |
Token | 'll |
Token ID | 1183 |
Feature activation | -9.45e-01 |
Pos logit contributions | |
supposed | +0.040 |
adders | +0.038 |
neys | +0.035 |
By | +0.035 |
emergencies | +0.034 |
Neg logit contributions | |
maid | -0.032 |
fish | -0.030 |
bl | -0.028 |
breath | -0.028 |
Qu | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | +7.45e-01 |
Pos logit contributions | |
supposed | +0.025 |
adders | +0.025 |
neys | +0.021 |
online | +0.021 |
emergencies | +0.020 |
Neg logit contributions | |
fish | -0.030 |
maid | -0.030 |
Qu | -0.030 |
ora | -0.030 |
bl | -0.027 |
Token | The |
Token ID | 464 |
Feature activation | +7.97e-01 |
Pos logit contributions | |
supposed | +0.029 |
adders | +0.029 |
neys | +0.025 |
online | +0.025 |
about | +0.024 |
Neg logit contributions | |
maid | -0.030 |
Qu | -0.029 |
fish | -0.029 |
ora | -0.028 |
bl | -0.027 |
Token | dad |
Token ID | 9955 |
Feature activation | +7.58e-01 |
Pos logit contributions | |
adders | +0.039 |
supposed | +0.036 |
neys | +0.035 |
By | +0.032 |
about | +0.031 |
Neg logit contributions | |
maid | -0.027 |
bl | -0.026 |
dolphin | -0.025 |
fish | -0.025 |
breath | -0.024 |
Token | tried |
Token ID | 3088 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
supposed | +0.032 |
adders | +0.032 |
neys | +0.028 |
emergencies | +0.027 |
By | +0.027 |
Neg logit contributions | |
maid | -0.029 |
fish | -0.028 |
bl | -0.026 |
Qu | -0.026 |
ora | -0.026 |
Token | to |
Token ID | 284 |
Feature activation | -4.87e-01 |
Pos logit contributions | |
breath | +0.090 |
bl | +0.089 |
maid | +0.089 |
fish | +0.085 |
snail | +0.084 |
Neg logit contributions | |
adders | -0.071 |
supposed | -0.063 |
neys | -0.060 |
eps | -0.051 |
online | -0.051 |
Token | help |
Token ID | 1037 |
Feature activation | -3.51e+00 |
Pos logit contributions | |
maid | +0.020 |
fish | +0.019 |
Qu | +0.018 |
ora | +0.018 |
bl | +0.017 |
Neg logit contributions | |
supposed | -0.019 |
adders | -0.019 |
neys | -0.016 |
measuring | -0.016 |
about | -0.016 |
Token | his |
Token ID | 465 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
maid | +0.142 |
fish | +0.134 |
bl | +0.128 |
ora | +0.128 |
breath | +0.127 |
Neg logit contributions | |
supposed | -0.142 |
adders | -0.138 |
neys | -0.122 |
By | -0.116 |
online | -0.115 |
Token | daughter |
Token ID | 4957 |
Feature activation | +9.98e-01 |
Pos logit contributions | |
maid | +0.030 |
bl | +0.028 |
breath | +0.027 |
ora | +0.026 |
dolphin | +0.025 |
Neg logit contributions | |
adders | -0.057 |
supposed | -0.051 |
neys | -0.051 |
By | -0.050 |
chairs | -0.047 |
Token | by |
Token ID | 416 |
Feature activation | +7.12e-01 |
Pos logit contributions | |
adders | +0.045 |
supposed | +0.043 |
neys | +0.041 |
emergencies | +0.040 |
By | +0.038 |
Neg logit contributions | |
bl | -0.032 |
breath | -0.032 |
maid | -0.030 |
fish | -0.027 |
snail | -0.027 |
Token | picking |
Token ID | 10868 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
supposed | +0.021 |
adders | +0.020 |
measuring | +0.017 |
about | +0.017 |
done | +0.016 |
Neg logit contributions | |
maid | -0.037 |
fish | -0.035 |
Qu | -0.034 |
ora | -0.034 |
bl | -0.032 |
Token | her |
Token ID | 607 |
Feature activation | -7.09e-01 |
Pos logit contributions | |
supposed | +0.013 |
adders | +0.012 |
neys | +0.011 |
By | +0.010 |
done | +0.010 |
Neg logit contributions | |
maid | -0.015 |
breath | -0.014 |
bl | -0.014 |
ora | -0.014 |
fish | -0.014 |