Token | . |
Token ID | 13 |
Feature activation | +2.50e-03 |
Pos logit contributions | |
arp | +0.033 |
cubes | +0.032 |
island | +0.029 |
bulbs | +0.027 |
cube | +0.026 |
Neg logit contributions | |
severe | -0.056 |
Deal | -0.055 |
amaz | -0.055 |
adays | -0.054 |
grin | -0.053 |
Token | Her |
Token ID | 2332 |
Feature activation | +5.43e-02 |
Pos logit contributions | |
arp | +0.004 |
island | +0.004 |
cubes | +0.004 |
bulbs | +0.004 |
place | +0.004 |
Neg logit contributions | |
grin | -0.003 |
amaz | -0.003 |
severe | -0.003 |
Deal | -0.003 |
opl | -0.003 |
Token | mom |
Token ID | 1995 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
cubes | +0.048 |
cube | +0.044 |
island | +0.040 |
arp | +0.040 |
collections | +0.040 |
Neg logit contributions | |
severe | -0.114 |
opl | -0.111 |
oled | -0.109 |
istent | -0.109 |
used | -0.105 |
Token | always |
Token ID | 1464 |
Feature activation | +7.51e-03 |
Pos logit contributions | |
cubes | +0.020 |
arp | +0.019 |
island | +0.019 |
bulbs | +0.017 |
cube | +0.016 |
Neg logit contributions | |
Deal | -0.060 |
amaz | -0.057 |
severe | -0.057 |
opl | -0.056 |
adays | -0.055 |
Token | said |
Token ID | 531 |
Feature activation | -5.00e-03 |
Pos logit contributions | |
cubes | +0.006 |
island | +0.005 |
cube | +0.005 |
ship | +0.005 |
arp | +0.004 |
Neg logit contributions | |
Deal | -0.016 |
adays | -0.016 |
opl | -0.016 |
severe | -0.015 |
amaz | -0.015 |
Token | Mell |
Token ID | 32993 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
opl | +0.011 |
Deal | +0.011 |
severe | +0.010 |
amaz | +0.009 |
used | +0.009 |
Neg logit contributions | |
cubes | -0.005 |
bulbs | -0.005 |
boats | -0.004 |
crystals | -0.004 |
shells | -0.004 |
Token | ie |
Token ID | 494 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
arp | +0.129 |
island | +0.108 |
cubes | +0.102 |
place | +0.088 |
bulbs | +0.087 |
Neg logit contributions | |
amaz | -0.317 |
grin | -0.316 |
Deal | -0.313 |
opl | -0.310 |
severe | -0.310 |
Token | was |
Token ID | 373 |
Feature activation | -3.98e-03 |
Pos logit contributions | |
bulbs | +0.013 |
arp | +0.013 |
cubes | +0.012 |
island | +0.012 |
creatures | +0.011 |
Neg logit contributions | |
Deal | -0.024 |
amaz | -0.022 |
severe | -0.022 |
opl | -0.022 |
grin | -0.021 |
Token | a |
Token ID | 257 |
Feature activation | -2.19e-03 |
Pos logit contributions | |
amaz | +0.007 |
severe | +0.007 |
adays | +0.007 |
opl | +0.007 |
grin | +0.007 |
Neg logit contributions | |
cubes | -0.005 |
bulbs | -0.004 |
island | -0.004 |
creatures | -0.004 |
boats | -0.004 |
Token | " |
Token ID | 366 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
opl | +0.003 |
severe | +0.003 |
adays | +0.003 |
Deal | +0.003 |
amaz | +0.003 |
Neg logit contributions | |
arp | -0.003 |
island | -0.003 |
cubes | -0.003 |
bulbs | -0.002 |
place | -0.002 |
Token | s |
Token ID | 82 |
Feature activation | +4.11e-03 |
Pos logit contributions | |
arp | +0.037 |
cubes | +0.031 |
place | +0.030 |
island | +0.029 |
bulbs | +0.028 |
Neg logit contributions | |
amaz | -0.052 |
Deal | -0.051 |
grin | -0.051 |
adays | -0.049 |
thank | -0.049 |
Token | so |
Token ID | 523 |
Feature activation | +9.13e-02 |
Pos logit contributions | |
island | +0.069 |
underground | +0.065 |
cubes | +0.063 |
creatures | +0.057 |
bulbs | +0.057 |
Neg logit contributions | |
adays | -0.131 |
oled | -0.130 |
severe | -0.125 |
opl | -0.124 |
Aud | -0.123 |
Token | surprised |
Token ID | 6655 |
Feature activation | +4.30e-02 |
Pos logit contributions | |
cubes | +0.072 |
organized | +0.070 |
cube | +0.065 |
pans | +0.061 |
creatures | +0.061 |
Neg logit contributions | |
severe | -0.197 |
hered | -0.195 |
oled | -0.188 |
Aud | -0.187 |
opl | -0.186 |
Token | and |
Token ID | 290 |
Feature activation | +5.63e-02 |
Pos logit contributions | |
arp | +0.055 |
island | +0.054 |
cubes | +0.050 |
place | +0.047 |
bulbs | +0.046 |
Neg logit contributions | |
Deal | -0.077 |
severe | -0.073 |
opl | -0.071 |
Aud | -0.070 |
oled | -0.069 |
Token | couldn |
Token ID | 3521 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
island | +0.049 |
creatures | +0.046 |
pans | +0.044 |
ships | +0.043 |
f | +0.043 |
Neg logit contributions | |
opl | -0.120 |
Deal | -0.112 |
hered | -0.106 |
oled | -0.106 |
severe | -0.106 |
Token | 't |
Token ID | 470 |
Feature activation | +3.56e-02 |
Pos logit contributions | |
arp | +0.012 |
guard | +0.010 |
place | +0.010 |
bulb | +0.010 |
Tim | +0.009 |
Neg logit contributions | |
amaz | -0.024 |
Majesty | -0.022 |
grin | -0.020 |
reflex | -0.020 |
Deal | -0.020 |
Token | believe |
Token ID | 1975 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
island | +0.050 |
arp | +0.049 |
place | +0.045 |
cubes | +0.041 |
bulbs | +0.040 |
Neg logit contributions | |
opl | -0.055 |
Aud | -0.054 |
Deal | -0.054 |
oled | -0.053 |
severe | -0.052 |
Token | it |
Token ID | 340 |
Feature activation | +5.95e-02 |
Pos logit contributions | |
arp | +0.244 |
island | +0.231 |
cubes | +0.227 |
bulbs | +0.212 |
place | +0.203 |
Neg logit contributions | |
Deal | -0.202 |
severe | -0.197 |
opl | -0.196 |
grin | -0.191 |
Aud | -0.190 |
Token | . |
Token ID | 13 |
Feature activation | +3.85e-02 |
Pos logit contributions | |
island | +0.052 |
arp | +0.048 |
place | +0.045 |
bulbs | +0.043 |
creatures | +0.041 |
Neg logit contributions | |
Deal | -0.127 |
severe | -0.123 |
opl | -0.122 |
grin | -0.121 |
icks | -0.119 |
Token | He |
Token ID | 679 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
arp | +0.051 |
cubes | +0.048 |
bulbs | +0.044 |
island | +0.044 |
Tim | +0.043 |
Neg logit contributions | |
severe | -0.056 |
opl | -0.055 |
icks | -0.053 |
grin | -0.053 |
amaz | -0.052 |
Token | said |
Token ID | 531 |
Feature activation | +5.68e-03 |
Pos logit contributions | |
island | +0.013 |
arp | +0.011 |
cubes | +0.010 |
ship | +0.010 |
place | +0.009 |
Neg logit contributions | |
opl | -0.032 |
Deal | -0.032 |
severe | -0.029 |
oled | -0.028 |
icks | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
cubes | +0.006 |
island | +0.006 |
cube | +0.006 |
shells | +0.005 |
bulbs | +0.005 |
Neg logit contributions | |
opl | -0.012 |
Deal | -0.012 |
istent | -0.010 |
Aud | -0.010 |
severe | -0.010 |
Token | the |
Token ID | 262 |
Feature activation | +5.68e-02 |
Pos logit contributions | |
island | +0.108 |
cubes | +0.107 |
crystals | +0.100 |
creatures | +0.099 |
bulbs | +0.097 |
Neg logit contributions | |
Deal | -0.180 |
severe | -0.180 |
oled | -0.178 |
istent | -0.175 |
ung | -0.168 |
Token | tide |
Token ID | 20013 |
Feature activation | +6.20e-02 |
Pos logit contributions | |
island | +0.063 |
arp | +0.056 |
cubes | +0.054 |
bulbs | +0.052 |
place | +0.051 |
Neg logit contributions | |
Deal | -0.111 |
oled | -0.111 |
istent | -0.109 |
icks | -0.109 |
severe | -0.108 |
Token | ." |
Token ID | 526 |
Feature activation | -3.38e-03 |
Pos logit contributions | |
island | +0.066 |
cubes | +0.055 |
arp | +0.053 |
place | +0.052 |
ship | +0.050 |
Neg logit contributions | |
Deal | -0.142 |
severe | -0.124 |
amaz | -0.120 |
Aud | -0.117 |
icks | -0.113 |
Token |
|
Token ID | 198 |
Feature activation | +3.56e-02 |
Pos logit contributions | |
amaz | +0.005 |
severe | +0.005 |
opl | +0.005 |
istent | +0.005 |
icks | +0.005 |
Neg logit contributions | |
arp | -0.004 |
cubes | -0.004 |
island | -0.004 |
Tim | -0.003 |
bulbs | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
arp | +0.057 |
island | +0.050 |
cubes | +0.049 |
place | +0.047 |
bulbs | +0.046 |
Neg logit contributions | |
amaz | -0.047 |
grin | -0.047 |
severe | -0.047 |
adays | -0.045 |
istent | -0.044 |
Token | Carl |
Token ID | 26886 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
arp | +0.032 |
island | +0.028 |
cubes | +0.027 |
place | +0.025 |
bulbs | +0.025 |
Neg logit contributions | |
amaz | -0.046 |
grin | -0.046 |
severe | -0.045 |
opl | -0.045 |
Deal | -0.045 |
Token | thought |
Token ID | 1807 |
Feature activation | +4.53e-02 |
Pos logit contributions | |
island | +0.119 |
arp | +0.117 |
cubes | +0.111 |
cube | +0.103 |
f | +0.094 |
Neg logit contributions | |
Deal | -0.304 |
Aud | -0.296 |
opl | -0.295 |
severe | -0.286 |
Paw | -0.274 |
Token | for |
Token ID | 329 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
cubes | +0.052 |
island | +0.047 |
arp | +0.047 |
pipes | +0.045 |
bulbs | +0.043 |
Neg logit contributions | |
opl | -0.080 |
severe | -0.079 |
Deal | -0.079 |
amaz | -0.077 |
Aud | -0.076 |
Token | a |
Token ID | 257 |
Feature activation | +8.66e-03 |
Pos logit contributions | |
grin | +0.016 |
opl | +0.016 |
amaz | +0.016 |
Deal | +0.016 |
severe | +0.015 |
Neg logit contributions | |
arp | -0.035 |
island | -0.033 |
cubes | -0.032 |
place | -0.031 |
bulbs | -0.031 |
Token | moment |
Token ID | 2589 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
arp | +0.008 |
island | +0.007 |
cubes | +0.007 |
place | +0.006 |
bulbs | +0.006 |
Neg logit contributions | |
opl | -0.017 |
amaz | -0.017 |
Deal | -0.016 |
grin | -0.016 |
severe | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
island | +0.039 |
arp | +0.038 |
cubes | +0.035 |
place | +0.033 |
bulbs | +0.031 |
Neg logit contributions | |
severe | -0.068 |
Deal | -0.068 |
opl | -0.065 |
Majesty | -0.065 |
adays | -0.064 |
Token | very |
Token ID | 845 |
Feature activation | +8.73e-02 |
Pos logit contributions | |
opl | +0.004 |
Deal | +0.004 |
severe | +0.004 |
adays | +0.004 |
istent | +0.004 |
Neg logit contributions | |
arp | -0.003 |
island | -0.003 |
cubes | -0.003 |
place | -0.003 |
bulbs | -0.003 |
Token | special |
Token ID | 2041 |
Feature activation | +2.09e-02 |
Pos logit contributions | |
arp | +0.103 |
island | +0.096 |
cubes | +0.096 |
place | +0.089 |
bulbs | +0.083 |
Neg logit contributions | |
severe | -0.151 |
opl | -0.151 |
oled | -0.151 |
Deal | -0.148 |
amaz | -0.148 |
Token | toy |
Token ID | 13373 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
arp | +0.027 |
island | +0.026 |
cubes | +0.025 |
bulbs | +0.023 |
place | +0.023 |
Neg logit contributions | |
Deal | -0.034 |
severe | -0.033 |
adays | -0.033 |
opl | -0.033 |
oled | -0.033 |
Token | for |
Token ID | 329 |
Feature activation | -4.01e-02 |
Pos logit contributions | |
arp | +0.038 |
island | +0.035 |
cubes | +0.034 |
bulbs | +0.032 |
place | +0.032 |
Neg logit contributions | |
grin | -0.032 |
amaz | -0.032 |
Deal | -0.031 |
thank | -0.031 |
opl | -0.031 |
Token | L |
Token ID | 406 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
severe | +0.056 |
grin | +0.056 |
Deal | +0.055 |
amaz | +0.055 |
opl | +0.055 |
Neg logit contributions | |
arp | -0.060 |
cubes | -0.060 |
island | -0.059 |
bulbs | -0.056 |
place | -0.054 |
Token | ott |
Token ID | 1252 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
arp | +0.012 |
island | +0.012 |
jackets | +0.010 |
cubes | +0.009 |
place | +0.009 |
Neg logit contributions | |
oled | -0.019 |
grin | -0.018 |
opl | -0.018 |
amaz | -0.018 |
Once | -0.018 |
Token | ie |
Token ID | 494 |
Feature activation | +1.74e-02 |
Pos logit contributions | |
arp | +0.145 |
island | +0.129 |
cubes | +0.123 |
place | +0.112 |
bulbs | +0.110 |
Neg logit contributions | |
opl | -0.280 |
grin | -0.267 |
severe | -0.263 |
amaz | -0.262 |
Deal | -0.261 |
Token | . |
Token ID | 13 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
arp | +0.020 |
island | +0.018 |
cubes | +0.018 |
place | +0.017 |
bulbs | +0.016 |
Neg logit contributions | |
Deal | -0.033 |
amaz | -0.032 |
severe | -0.031 |
grin | -0.031 |
opl | -0.031 |
Token | It |
Token ID | 632 |
Feature activation | +3.95e-02 |
Pos logit contributions | |
arp | +0.037 |
cubes | +0.034 |
bulbs | +0.032 |
island | +0.032 |
pipes | +0.031 |
Neg logit contributions | |
opl | -0.023 |
grin | -0.022 |
Deal | -0.022 |
amaz | -0.021 |
severe | -0.021 |
Token | was |
Token ID | 373 |
Feature activation | +3.90e-04 |
Pos logit contributions | |
arp | +0.032 |
island | +0.030 |
bulbs | +0.028 |
cubes | +0.026 |
place | +0.023 |
Neg logit contributions | |
Deal | -0.091 |
severe | -0.086 |
opl | -0.085 |
amaz | -0.083 |
grin | -0.082 |
Token | a |
Token ID | 257 |
Feature activation | +2.88e-05 |
Pos logit contributions | |
cubes | +0.000 |
bulbs | +0.000 |
island | +0.000 |
crystals | +0.000 |
arp | +0.000 |
Neg logit contributions | |
severe | -0.001 |
used | -0.001 |
Deal | -0.001 |
opl | -0.001 |
istent | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
Deal | +0.004 |
severe | +0.004 |
grin | +0.004 |
amaz | +0.004 |
opl | +0.004 |
Neg logit contributions | |
island | -0.003 |
arp | -0.003 |
place | -0.003 |
cubes | -0.003 |
bulbs | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | +5.93e-05 |
Pos logit contributions | |
arp | +0.022 |
island | +0.020 |
cubes | +0.019 |
bulbs | +0.018 |
place | +0.018 |
Neg logit contributions | |
amaz | -0.014 |
grin | -0.014 |
opl | -0.013 |
severe | -0.013 |
Deal | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | +2.79e-03 |
Pos logit contributions | |
arp | +0.000 |
island | +0.000 |
cubes | +0.000 |
bulbs | +0.000 |
place | +0.000 |
Neg logit contributions | |
grin | -0.000 |
Once | -0.000 |
amaz | -0.000 |
Deal | -0.000 |
Immediately | -0.000 |
Token | When |
Token ID | 2215 |
Feature activation | -2.70e-02 |
Pos logit contributions | |
arp | +0.005 |
island | +0.005 |
cubes | +0.004 |
bulbs | +0.004 |
crystals | +0.004 |
Neg logit contributions | |
Once | -0.003 |
adays | -0.003 |
Immediately | -0.003 |
oled | -0.003 |
Aud | -0.003 |
Token | D |
Token ID | 360 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
thank | +0.024 |
bye | +0.023 |
adays | +0.022 |
Once | +0.022 |
Mitt | +0.022 |
Neg logit contributions | |
arp | -0.057 |
island | -0.052 |
cubes | -0.051 |
place | -0.050 |
bulbs | -0.048 |
Token | ott |
Token ID | 1252 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
island | +0.011 |
arp | +0.010 |
cubes | +0.009 |
jackets | +0.008 |
bulbs | +0.008 |
Neg logit contributions | |
opl | -0.020 |
grin | -0.020 |
Once | -0.019 |
amaz | -0.019 |
Deal | -0.019 |
Token | ie |
Token ID | 494 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
arp | +0.156 |
cubes | +0.137 |
island | +0.136 |
place | +0.128 |
bulbs | +0.124 |
Neg logit contributions | |
opl | -0.270 |
amaz | -0.252 |
severe | -0.250 |
grin | -0.249 |
zing | -0.248 |
Token | arrived |
Token ID | 5284 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
island | +0.025 |
arp | +0.024 |
place | +0.022 |
ship | +0.019 |
cubes | +0.019 |
Neg logit contributions | |
Deal | -0.037 |
opl | -0.034 |
grin | -0.034 |
amaz | -0.034 |
severe | -0.034 |
Token | at |
Token ID | 379 |
Feature activation | +1.35e-03 |
Pos logit contributions | |
arp | +0.060 |
island | +0.057 |
cubes | +0.057 |
bulbs | +0.053 |
place | +0.052 |
Neg logit contributions | |
opl | -0.041 |
amaz | -0.040 |
Deal | -0.040 |
severe | -0.038 |
grin | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
arp | +0.002 |
island | +0.002 |
cubes | +0.002 |
bulbs | +0.002 |
place | +0.002 |
Neg logit contributions | |
grin | -0.002 |
istent | -0.002 |
Deal | -0.002 |
oled | -0.002 |
severe | -0.002 |
Token | lab |
Token ID | 2248 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
arp | +0.020 |
cubes | +0.018 |
island | +0.018 |
bulbs | +0.017 |
place | +0.015 |
Neg logit contributions | |
amaz | -0.023 |
Deal | -0.023 |
thank | -0.022 |
opl | -0.022 |
grin | -0.022 |
Token | oven |
Token ID | 14361 |
Feature activation | -3.73e-02 |
Pos logit contributions | |
arp | +0.024 |
island | +0.022 |
cubes | +0.021 |
bulbs | +0.020 |
place | +0.019 |
Neg logit contributions | |
amaz | -0.022 |
grin | -0.022 |
severe | -0.021 |
Deal | -0.021 |
opl | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
Deal | +0.078 |
severe | +0.077 |
grin | +0.075 |
Aud | +0.075 |
opl | +0.075 |
Neg logit contributions | |
arp | -0.034 |
island | -0.033 |
cubes | -0.030 |
place | -0.026 |
bulbs | -0.026 |
Token | |
Token ID | 220 |
Feature activation | -8.13e-03 |
Pos logit contributions | |
arp | +0.054 |
island | +0.049 |
cubes | +0.049 |
bulbs | +0.047 |
place | +0.045 |
Neg logit contributions | |
opl | -0.032 |
amaz | -0.030 |
grin | -0.030 |
severe | -0.028 |
Deal | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | +2.47e-02 |
Pos logit contributions | |
amaz | +0.014 |
grin | +0.014 |
severe | +0.014 |
bye | +0.013 |
opl | +0.013 |
Neg logit contributions | |
arp | -0.010 |
cubes | -0.008 |
bulbs | -0.008 |
place | -0.007 |
leaning | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | +2.42e-02 |
Pos logit contributions | |
arp | +0.045 |
island | +0.042 |
cubes | +0.041 |
bulbs | +0.039 |
place | +0.039 |
Neg logit contributions | |
amaz | -0.027 |
grin | -0.026 |
Deal | -0.025 |
opl | -0.025 |
severe | -0.025 |
Token | P |
Token ID | 47 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
arp | +0.038 |
island | +0.037 |
cubes | +0.036 |
bulbs | +0.033 |
place | +0.031 |
Neg logit contributions | |
adays | -0.033 |
Deal | -0.033 |
severe | -0.032 |
oled | -0.032 |
Immediately | -0.032 |
Token | enny |
Token ID | 11870 |
Feature activation | +4.90e-02 |
Pos logit contributions | |
arp | +0.230 |
island | +0.202 |
cubes | +0.200 |
place | +0.197 |
bulbs | +0.185 |
Neg logit contributions | |
amaz | -0.202 |
grin | -0.201 |
opl | -0.199 |
Deal | -0.189 |
thank | -0.188 |
Token | and |
Token ID | 290 |
Feature activation | +2.27e-02 |
Pos logit contributions | |
arp | +0.061 |
island | +0.057 |
cubes | +0.053 |
bulbs | +0.049 |
place | +0.048 |
Neg logit contributions | |
Deal | -0.084 |
opl | -0.083 |
severe | -0.081 |
amaz | -0.081 |
grin | -0.078 |
Token | her |
Token ID | 607 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
arp | +0.043 |
island | +0.040 |
cubes | +0.040 |
bulbs | +0.038 |
place | +0.037 |
Neg logit contributions | |
amaz | -0.022 |
opl | -0.022 |
grin | -0.022 |
Deal | -0.022 |
severe | -0.021 |
Token | mom |
Token ID | 1995 |
Feature activation | +4.49e-02 |
Pos logit contributions | |
arp | +0.041 |
island | +0.037 |
cubes | +0.037 |
bulbs | +0.034 |
place | +0.034 |
Neg logit contributions | |
amaz | -0.036 |
grin | -0.035 |
severe | -0.035 |
Deal | -0.035 |
opl | -0.035 |
Token | my |
Token ID | 1820 |
Feature activation | -3.82e-02 |
Pos logit contributions | |
arp | +0.069 |
island | +0.062 |
cubes | +0.061 |
bulbs | +0.057 |
place | +0.057 |
Neg logit contributions | |
amaz | -0.062 |
grin | -0.062 |
Deal | -0.061 |
severe | -0.059 |
opl | -0.059 |
Token | just |
Token ID | 655 |
Feature activation | +4.46e-02 |
Pos logit contributions | |
place | +0.099 |
island | +0.099 |
treasures | +0.095 |
creatures | +0.094 |
bulbs | +0.091 |
Neg logit contributions | |
Deal | -0.163 |
oled | -0.147 |
opl | -0.146 |
istent | -0.145 |
severe | -0.140 |
Token | begun |
Token ID | 9258 |
Feature activation | +2.87e-02 |
Pos logit contributions | |
island | +0.052 |
place | +0.051 |
creatures | +0.049 |
treasures | +0.047 |
shells | +0.046 |
Neg logit contributions | |
oled | -0.080 |
Deal | -0.080 |
Aud | -0.077 |
Sean | -0.073 |
istent | -0.070 |
Token | ! |
Token ID | 0 |
Feature activation | -5.13e-04 |
Pos logit contributions | |
bulbs | +0.028 |
island | +0.028 |
creatures | +0.027 |
place | +0.027 |
treasures | +0.026 |
Neg logit contributions | |
Deal | -0.060 |
oled | -0.058 |
istent | -0.054 |
grin | -0.054 |
opl | -0.054 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +8.36e-02 |
Pos logit contributions | |
grin | +0.001 |
oled | +0.001 |
severe | +0.001 |
amaz | +0.001 |
istent | +0.001 |
Neg logit contributions | |
arp | -0.000 |
island | -0.000 |
B | -0.000 |
bulbs | -0.000 |
ney | -0.000 |
Token | F |
Token ID | 37 |
Feature activation | +3.94e-02 |
Pos logit contributions | |
arp | +0.090 |
island | +0.080 |
cubes | +0.079 |
bulbs | +0.071 |
place | +0.068 |
Neg logit contributions | |
amaz | -0.151 |
severe | -0.148 |
grin | -0.148 |
Deal | -0.147 |
istent | -0.147 |
Token | inn |
Token ID | 3732 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
arp | +0.046 |
island | +0.041 |
cubes | +0.038 |
place | +0.036 |
bulbs | +0.035 |
Neg logit contributions | |
amaz | -0.069 |
grin | -0.067 |
severe | -0.066 |
Deal | -0.066 |
Aud | -0.065 |
Token | was |
Token ID | 373 |
Feature activation | +7.01e-03 |
Pos logit contributions | |
arp | +0.137 |
island | +0.129 |
cubes | +0.128 |
bulbs | +0.104 |
pipes | +0.098 |
Neg logit contributions | |
Deal | -0.313 |
amaz | -0.307 |
Aud | -0.296 |
grin | -0.295 |
severe | -0.290 |
Token | fishing |
Token ID | 12478 |
Feature activation | +4.30e-02 |
Pos logit contributions | |
underground | +0.008 |
island | +0.007 |
cubes | +0.007 |
P | +0.007 |
bulbs | +0.007 |
Neg logit contributions | |
adays | -0.012 |
severe | -0.012 |
amaz | -0.011 |
grin | -0.011 |
thank | -0.011 |
Token | with |
Token ID | 351 |
Feature activation | +6.61e-03 |
Pos logit contributions | |
arp | +0.051 |
island | +0.051 |
cubes | +0.050 |
place | +0.046 |
cube | +0.041 |
Neg logit contributions | |
Deal | -0.078 |
amaz | -0.077 |
oled | -0.075 |
Aud | -0.075 |
grin | -0.075 |
Token | his |
Token ID | 465 |
Feature activation | +4.22e-02 |
Pos logit contributions | |
cubes | +0.008 |
shells | +0.007 |
island | +0.007 |
crystals | +0.007 |
boats | +0.007 |
Neg logit contributions | |
oled | -0.013 |
istent | -0.013 |
severe | -0.013 |
adays | -0.012 |
grin | -0.012 |
Token | dad |
Token ID | 9955 |
Feature activation | +2.09e-02 |
Pos logit contributions | |
cubes | +0.049 |
island | +0.045 |
arp | +0.042 |
cube | +0.041 |
boats | +0.039 |
Neg logit contributions | |
oled | -0.083 |
adays | -0.080 |
severe | -0.080 |
istent | -0.079 |
opl | -0.078 |
Token | was |
Token ID | 373 |
Feature activation | +6.92e-02 |
Pos logit contributions | |
arp | +0.032 |
island | +0.030 |
cubes | +0.029 |
place | +0.028 |
bulbs | +0.028 |
Neg logit contributions | |
amaz | -0.020 |
Deal | -0.019 |
opl | -0.019 |
Aud | -0.018 |
severe | -0.018 |
Token | seeing |
Token ID | 4379 |
Feature activation | +8.28e-02 |
Pos logit contributions | |
island | +0.069 |
underground | +0.065 |
cubes | +0.063 |
creatures | +0.057 |
bulbs | +0.057 |
Neg logit contributions | |
adays | -0.131 |
oled | -0.130 |
severe | -0.125 |
opl | -0.124 |
Aud | -0.123 |
Token | . |
Token ID | 13 |
Feature activation | +6.66e-02 |
Pos logit contributions | |
island | +0.101 |
cubes | +0.096 |
creatures | +0.091 |
shells | +0.090 |
bulbs | +0.088 |
Neg logit contributions | |
Deal | -0.170 |
oled | -0.157 |
severe | -0.156 |
istent | -0.156 |
Aud | -0.152 |
Token |
|
Token ID | 198 |
Feature activation | +3.48e-02 |
Pos logit contributions | |
arp | +0.081 |
cubes | +0.080 |
bulbs | +0.072 |
island | +0.071 |
Tim | +0.071 |
Neg logit contributions | |
opl | -0.098 |
severe | -0.097 |
amaz | -0.094 |
grin | -0.094 |
zing | -0.094 |
Token |
|
Token ID | 198 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
arp | +0.050 |
cubes | +0.043 |
island | +0.042 |
place | +0.042 |
B | +0.040 |
Neg logit contributions | |
grin | -0.051 |
severe | -0.051 |
amaz | -0.050 |
adays | -0.049 |
ishly | -0.049 |
Token | P |
Token ID | 47 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
arp | +0.028 |
cubes | +0.024 |
island | +0.024 |
place | +0.023 |
leaning | +0.021 |
Neg logit contributions | |
grin | -0.037 |
amaz | -0.037 |
thank | -0.036 |
severe | -0.035 |
opl | -0.035 |
Token | olly |
Token ID | 5098 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
arp | +0.192 |
cubes | +0.184 |
place | +0.175 |
island | +0.167 |
collections | +0.162 |
Neg logit contributions | |
amaz | -0.221 |
severe | -0.217 |
oled | -0.216 |
opl | -0.214 |
grin | -0.214 |
Token | wanted |
Token ID | 2227 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
island | +0.017 |
arp | +0.014 |
ship | +0.013 |
boy | +0.013 |
creatures | +0.012 |
Neg logit contributions | |
opl | -0.043 |
Deal | -0.041 |
severe | -0.039 |
icks | -0.038 |
upid | -0.037 |
Token | to |
Token ID | 284 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
arp | +0.031 |
island | +0.030 |
cubes | +0.030 |
place | +0.028 |
bulbs | +0.027 |
Neg logit contributions | |
Deal | -0.050 |
opl | -0.049 |
severe | -0.049 |
oled | -0.048 |
grin | -0.047 |
Token | find |
Token ID | 1064 |
Feature activation | +7.13e-02 |
Pos logit contributions | |
island | +0.037 |
arp | +0.035 |
place | +0.034 |
cubes | +0.032 |
cube | +0.030 |
Neg logit contributions | |
oled | -0.054 |
Deal | -0.047 |
Aud | -0.046 |
severe | -0.046 |
opl | -0.046 |
Token | out |
Token ID | 503 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
arp | +0.124 |
island | +0.115 |
cubes | +0.112 |
bulbs | +0.111 |
place | +0.108 |
Neg logit contributions | |
grin | -0.089 |
oled | -0.089 |
Deal | -0.087 |
opl | -0.087 |
istent | -0.086 |
Token | . |
Token ID | 13 |
Feature activation | +2.85e-02 |
Pos logit contributions | |
Deal | +0.017 |
amaz | +0.017 |
grin | +0.016 |
Aud | +0.016 |
severe | +0.016 |
Neg logit contributions | |
arp | -0.010 |
cubes | -0.009 |
island | -0.009 |
bulbs | -0.008 |
pipes | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
arp | +0.043 |
cubes | +0.043 |
island | +0.039 |
place | +0.037 |
leaning | +0.036 |
Neg logit contributions | |
severe | -0.037 |
amaz | -0.036 |
opl | -0.036 |
grin | -0.035 |
used | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
arp | +0.036 |
island | +0.033 |
cubes | +0.033 |
bulbs | +0.031 |
place | +0.031 |
Neg logit contributions | |
amaz | -0.019 |
grin | -0.019 |
Deal | -0.018 |
severe | -0.018 |
opl | -0.018 |
Token | " |
Token ID | 1 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
arp | +0.032 |
island | +0.029 |
cubes | +0.028 |
bulbs | +0.027 |
place | +0.026 |
Neg logit contributions | |
amaz | -0.026 |
grin | -0.026 |
Deal | -0.025 |
severe | -0.025 |
opl | -0.025 |
Token | C |
Token ID | 34 |
Feature activation | +2.99e-02 |
Pos logit contributions | |
arp | +0.034 |
island | +0.033 |
cubes | +0.029 |
bulbs | +0.029 |
collections | +0.028 |
Neg logit contributions | |
Aud | -0.033 |
Deal | -0.033 |
amaz | -0.032 |
oled | -0.032 |
adays | -0.031 |
Token | leaning |
Token ID | 25909 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
cubes | +0.040 |
arp | +0.038 |
island | +0.038 |
bulbs | +0.037 |
crystals | +0.036 |
Neg logit contributions | |
opl | -0.043 |
severe | -0.041 |
adays | -0.040 |
Aud | -0.039 |
istent | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
arp | +0.234 |
island | +0.198 |
cubes | +0.198 |
bulbs | +0.195 |
pipes | +0.184 |
Neg logit contributions | |
amaz | -0.233 |
grin | -0.232 |
Aud | -0.229 |
Deal | -0.228 |
oled | -0.219 |
Token | sheep |
Token ID | 15900 |
Feature activation | -6.29e-03 |
Pos logit contributions | |
arp | +0.021 |
island | +0.019 |
bulbs | +0.016 |
pipes | +0.015 |
place | +0.015 |
Neg logit contributions | |
grin | -0.042 |
amaz | -0.041 |
Deal | -0.040 |
adays | -0.040 |
severe | -0.040 |
Token | is |
Token ID | 318 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
Deal | +0.012 |
amaz | +0.012 |
Aud | +0.012 |
grin | +0.012 |
severe | +0.011 |
Neg logit contributions | |
arp | -0.007 |
island | -0.007 |
cubes | -0.006 |
bulbs | -0.006 |
place | -0.006 |
Token | fun |
Token ID | 1257 |
Feature activation | +6.04e-02 |
Pos logit contributions | |
amaz | +0.042 |
grin | +0.042 |
Deal | +0.041 |
severe | +0.041 |
opl | +0.041 |
Neg logit contributions | |
arp | -0.036 |
cubes | -0.033 |
island | -0.033 |
bulbs | -0.031 |
place | -0.030 |
Token | !" |
Token ID | 2474 |
Feature activation | -4.30e-02 |
Pos logit contributions | |
arp | +0.106 |
island | +0.099 |
cubes | +0.097 |
bulbs | +0.092 |
place | +0.091 |
Neg logit contributions | |
amaz | -0.071 |
Deal | -0.071 |
grin | -0.070 |
severe | -0.069 |
opl | -0.067 |
Token | and |
Token ID | 290 |
Feature activation | +5.02e-03 |
Pos logit contributions | |
island | +0.021 |
arp | +0.020 |
cubes | +0.018 |
tunnel | +0.015 |
ship | +0.015 |
Neg logit contributions | |
Deal | -0.051 |
severe | -0.048 |
Aud | -0.044 |
opl | -0.043 |
Majesty | -0.042 |
Token | started |
Token ID | 2067 |
Feature activation | +1.93e-02 |
Pos logit contributions | |
arp | +0.005 |
bulbs | +0.005 |
island | +0.005 |
creatures | +0.005 |
crystals | +0.005 |
Neg logit contributions | |
opl | -0.010 |
Deal | -0.010 |
urance | -0.009 |
adays | -0.009 |
istent | -0.009 |
Token | leaping |
Token ID | 45583 |
Feature activation | -6.32e-02 |
Pos logit contributions | |
cubes | +0.016 |
pipes | +0.016 |
creatures | +0.015 |
f | +0.014 |
bulbs | +0.014 |
Neg logit contributions | |
Deal | -0.042 |
opl | -0.039 |
adays | -0.039 |
severe | -0.038 |
istent | -0.037 |
Token | around |
Token ID | 1088 |
Feature activation | +2.51e-03 |
Pos logit contributions | |
Deal | +0.135 |
opl | +0.127 |
Majesty | +0.122 |
istent | +0.121 |
grin | +0.118 |
Neg logit contributions | |
arp | -0.064 |
underground | -0.055 |
island | -0.054 |
tunnels | -0.054 |
bulbs | -0.051 |
Token | ! |
Token ID | 0 |
Feature activation | -4.30e-02 |
Pos logit contributions | |
arp | +0.003 |
bulbs | +0.003 |
island | +0.002 |
creatures | +0.002 |
cubes | +0.002 |
Neg logit contributions | |
Deal | -0.005 |
severe | -0.005 |
opl | -0.005 |
istent | -0.005 |
grin | -0.005 |
Token | Mell |
Token ID | 32993 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
opl | +0.062 |
Aud | +0.055 |
severe | +0.054 |
Deal | +0.052 |
grin | +0.052 |
Neg logit contributions | |
arp | -0.066 |
bulbs | -0.062 |
pipes | -0.055 |
crystals | -0.054 |
island | -0.053 |
Token | ie |
Token ID | 494 |
Feature activation | +4.30e-03 |
Pos logit contributions | |
arp | +0.141 |
island | +0.122 |
cubes | +0.121 |
place | +0.103 |
bulbs | +0.097 |
Neg logit contributions | |
grin | -0.277 |
amaz | -0.277 |
adays | -0.268 |
Deal | -0.267 |
oled | -0.264 |
Token | covered |
Token ID | 5017 |
Feature activation | +4.59e-02 |
Pos logit contributions | |
arp | +0.003 |
island | +0.003 |
creatures | +0.002 |
tunnel | +0.002 |
B | +0.002 |
Neg logit contributions | |
Deal | -0.010 |
opl | -0.010 |
severe | -0.009 |
Aud | -0.009 |
amaz | -0.009 |
Token | her |
Token ID | 607 |
Feature activation | +7.21e-03 |
Pos logit contributions | |
arp | +0.068 |
bulbs | +0.066 |
island | +0.063 |
cubes | +0.060 |
crystals | +0.059 |
Neg logit contributions | |
opl | -0.071 |
Deal | -0.069 |
severe | -0.068 |
Aud | -0.067 |
grin | -0.064 |
Token | eyes |
Token ID | 2951 |
Feature activation | -3.02e-02 |
Pos logit contributions | |
island | +0.008 |
cubes | +0.007 |
cube | +0.007 |
arp | +0.007 |
tunnel | +0.007 |
Neg logit contributions | |
oled | -0.013 |
Aud | -0.013 |
opl | -0.013 |
severe | -0.013 |
Deal | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
Deal | +0.075 |
Majesty | +0.069 |
opl | +0.066 |
majesty | +0.062 |
Jeffrey | +0.062 |
Neg logit contributions | |
island | -0.019 |
arp | -0.018 |
underground | -0.016 |
cubes | -0.016 |
in | -0.015 |
Token | gun |
Token ID | 2485 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
arp | +0.043 |
cubes | +0.041 |
island | +0.037 |
bulbs | +0.037 |
pipes | +0.034 |
Neg logit contributions | |
Deal | -0.076 |
grin | -0.070 |
severe | -0.069 |
oled | -0.069 |
used | -0.069 |
Token | ! |
Token ID | 0 |
Feature activation | -1.63e-03 |
Pos logit contributions | |
island | +0.040 |
arp | +0.040 |
cubes | +0.037 |
ship | +0.035 |
place | +0.033 |
Neg logit contributions | |
Deal | -0.080 |
amaz | -0.079 |
Aud | -0.072 |
grin | -0.071 |
delight | -0.070 |
Token | He |
Token ID | 679 |
Feature activation | +2.90e-02 |
Pos logit contributions | |
amaz | +0.002 |
Aud | +0.002 |
grin | +0.002 |
zing | +0.002 |
opl | +0.002 |
Neg logit contributions | |
arp | -0.002 |
cubes | -0.002 |
Tim | -0.002 |
cube | -0.002 |
island | -0.002 |
Token | couldn |
Token ID | 3521 |
Feature activation | +7.29e-03 |
Pos logit contributions | |
island | +0.025 |
ship | +0.023 |
ships | +0.021 |
boats | +0.019 |
arp | +0.019 |
Neg logit contributions | |
Deal | -0.064 |
Aud | -0.059 |
opl | -0.059 |
� | -0.057 |
oled | -0.056 |
Token | 't |
Token ID | 470 |
Feature activation | +3.10e-02 |
Pos logit contributions | |
arp | +0.008 |
guard | +0.007 |
place | +0.007 |
cube | +0.007 |
Tim | +0.007 |
Neg logit contributions | |
amaz | -0.015 |
Majesty | -0.013 |
Deal | -0.013 |
grin | -0.013 |
fascination | -0.012 |
Token | believe |
Token ID | 1975 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
arp | +0.053 |
island | +0.050 |
cubes | +0.047 |
place | +0.045 |
bulbs | +0.044 |
Neg logit contributions | |
amaz | -0.038 |
Deal | -0.037 |
grin | -0.037 |
opl | -0.036 |
severe | -0.036 |
Token | his |
Token ID | 465 |
Feature activation | +4.96e-02 |
Pos logit contributions | |
arp | +0.224 |
island | +0.214 |
cubes | +0.212 |
bulbs | +0.192 |
crystals | +0.187 |
Neg logit contributions | |
Deal | -0.206 |
severe | -0.197 |
opl | -0.197 |
Aud | -0.196 |
grin | -0.192 |
Token | eyes |
Token ID | 2951 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
island | +0.070 |
arp | +0.069 |
cubes | +0.067 |
ship | +0.062 |
anchor | +0.061 |
Neg logit contributions | |
Aud | -0.085 |
oled | -0.083 |
Ashley | -0.076 |
Immediately | -0.074 |
Heidi | -0.073 |
Token | . |
Token ID | 13 |
Feature activation | +3.10e-02 |
Pos logit contributions | |
Deal | +0.036 |
Aud | +0.031 |
Majesty | +0.029 |
Elizabeth | +0.029 |
Sean | +0.029 |
Neg logit contributions | |
island | -0.015 |
arp | -0.012 |
cubes | -0.011 |
underground | -0.011 |
place | -0.011 |
Token | He |
Token ID | 679 |
Feature activation | +6.16e-03 |
Pos logit contributions | |
arp | +0.038 |
cubes | +0.036 |
island | +0.032 |
bulbs | +0.032 |
Tim | +0.031 |
Neg logit contributions | |
severe | -0.047 |
grin | -0.047 |
amaz | -0.046 |
Aud | -0.046 |
icks | -0.045 |
Token | had |
Token ID | 550 |
Feature activation | +5.52e-02 |
Pos logit contributions | |
island | +0.006 |
ship | +0.004 |
arp | +0.004 |
ships | +0.004 |
boats | +0.003 |
Neg logit contributions | |
Deal | -0.014 |
opl | -0.013 |
Aud | -0.013 |
severe | -0.012 |
icks | -0.012 |
Token | so |
Token ID | 523 |
Feature activation | +9.13e-02 |
Pos logit contributions | |
island | +0.069 |
underground | +0.065 |
cubes | +0.063 |
creatures | +0.057 |
bulbs | +0.057 |
Neg logit contributions | |
adays | -0.131 |
oled | -0.130 |
severe | -0.125 |
opl | -0.124 |
Aud | -0.123 |
Token | cool |
Token ID | 3608 |
Feature activation | +6.40e-02 |
Pos logit contributions | |
cubes | +0.072 |
organized | +0.070 |
cube | +0.065 |
pans | +0.061 |
creatures | +0.061 |
Neg logit contributions | |
severe | -0.197 |
hered | -0.195 |
oled | -0.188 |
Aud | -0.187 |
opl | -0.186 |
Token | and |
Token ID | 290 |
Feature activation | +5.11e-02 |
Pos logit contributions | |
island | +0.082 |
arp | +0.077 |
cubes | +0.072 |
bulbs | +0.064 |
creatures | +0.064 |
Neg logit contributions | |
Deal | -0.127 |
severe | -0.109 |
amaz | -0.108 |
opl | -0.107 |
icks | -0.106 |
Token | couldn |
Token ID | 3521 |
Feature activation | +4.56e-03 |
Pos logit contributions | |
island | +0.047 |
cubes | +0.047 |
creatures | +0.045 |
shells | +0.044 |
treasures | +0.042 |
Neg logit contributions | |
Deal | -0.108 |
severe | -0.103 |
oled | -0.103 |
opl | -0.102 |
hered | -0.099 |
Token | 't |
Token ID | 470 |
Feature activation | +3.11e-02 |
Pos logit contributions | |
arp | +0.005 |
place | +0.004 |
guard | +0.004 |
island | +0.004 |
bulb | +0.004 |
Neg logit contributions | |
amaz | -0.009 |
Deal | -0.008 |
Majesty | -0.008 |
grin | -0.008 |
reflex | -0.007 |
Token | believe |
Token ID | 1975 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
arp | +0.044 |
island | +0.044 |
place | +0.039 |
cubes | +0.037 |
bulbs | +0.036 |
Neg logit contributions | |
Deal | -0.047 |
amaz | -0.046 |
Aud | -0.046 |
opl | -0.046 |
oled | -0.046 |
Token | what |
Token ID | 644 |
Feature activation | +6.05e-02 |
Pos logit contributions | |
arp | +0.227 |
island | +0.216 |
cubes | +0.212 |
bulbs | +0.197 |
crystals | +0.189 |
Neg logit contributions | |
Deal | -0.200 |
severe | -0.195 |
opl | -0.194 |
grin | -0.189 |
Aud | -0.188 |
Token | he |
Token ID | 339 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
cubes | +0.072 |
island | +0.071 |
bulbs | +0.066 |
place | +0.066 |
crystals | +0.065 |
Neg logit contributions | |
Deal | -0.109 |
severe | -0.107 |
opl | -0.106 |
oled | -0.101 |
istent | -0.101 |
Token | had |
Token ID | 550 |
Feature activation | +5.93e-02 |
Pos logit contributions | |
island | +0.012 |
arp | +0.010 |
place | +0.010 |
cubes | +0.008 |
ship | +0.007 |
Neg logit contributions | |
Deal | -0.036 |
opl | -0.035 |
severe | -0.035 |
adays | -0.034 |
icks | -0.033 |
Token | found |
Token ID | 1043 |
Feature activation | +6.59e-02 |
Pos logit contributions | |
island | +0.068 |
arp | +0.065 |
cubes | +0.063 |
place | +0.057 |
collections | +0.053 |
Neg logit contributions | |
Deal | -0.111 |
severe | -0.104 |
grin | -0.102 |
opl | -0.101 |
Aud | -0.100 |
Token | . |
Token ID | 13 |
Feature activation | +3.47e-02 |
Pos logit contributions | |
island | +0.067 |
arp | +0.067 |
cubes | +0.063 |
place | +0.059 |
crystals | +0.054 |
Neg logit contributions | |
Deal | -0.141 |
severe | -0.130 |
Aud | -0.129 |
grin | -0.126 |
istent | -0.122 |
Token | believe |
Token ID | 1975 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
island | +0.100 |
arp | +0.098 |
place | +0.090 |
cubes | +0.084 |
bulbs | +0.082 |
Neg logit contributions | |
Deal | -0.098 |
amaz | -0.097 |
oled | -0.095 |
Aud | -0.095 |
adays | -0.091 |
Token | it |
Token ID | 340 |
Feature activation | +6.84e-02 |
Pos logit contributions | |
arp | +0.256 |
island | +0.253 |
cubes | +0.248 |
bulbs | +0.229 |
crystals | +0.222 |
Neg logit contributions | |
Deal | -0.254 |
severe | -0.249 |
opl | -0.246 |
amaz | -0.242 |
oled | -0.239 |
Token | ! |
Token ID | 0 |
Feature activation | -1.35e-03 |
Pos logit contributions | |
island | +0.054 |
place | +0.045 |
creatures | +0.040 |
arp | +0.039 |
bulbs | +0.037 |
Neg logit contributions | |
Deal | -0.157 |
icks | -0.148 |
Aud | -0.147 |
opl | -0.145 |
severe | -0.144 |
Token |
|
Token ID | 198 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
opl | +0.002 |
severe | +0.002 |
amaz | +0.002 |
grin | +0.002 |
zing | +0.002 |
Neg logit contributions | |
arp | -0.002 |
bulbs | -0.002 |
cubes | -0.002 |
island | -0.001 |
Tim | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
arp | +0.054 |
island | +0.046 |
place | +0.045 |
cubes | +0.045 |
B | +0.041 |
Neg logit contributions | |
severe | -0.056 |
amaz | -0.055 |
grin | -0.055 |
adays | -0.053 |
Deal | -0.053 |
Token | P |
Token ID | 47 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
arp | +0.035 |
island | +0.030 |
place | +0.029 |
cubes | +0.028 |
B | +0.026 |
Neg logit contributions | |
grin | -0.049 |
amaz | -0.049 |
thank | -0.048 |
severe | -0.048 |
icks | -0.048 |
Token | apa |
Token ID | 32678 |
Feature activation | -8.17e-03 |
Pos logit contributions | |
arp | +0.211 |
place | +0.197 |
cubes | +0.196 |
island | +0.192 |
collections | +0.181 |
Neg logit contributions | |
amaz | -0.200 |
severe | -0.196 |
oled | -0.195 |
opl | -0.193 |
grin | -0.192 |
Token | explained |
Token ID | 4893 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
Deal | +0.017 |
opl | +0.016 |
amaz | +0.016 |
severe | +0.016 |
Aud | +0.015 |
Neg logit contributions | |
island | -0.008 |
arp | -0.006 |
ship | -0.006 |
cubes | -0.006 |
boy | -0.006 |
Token | that |
Token ID | 326 |
Feature activation | +9.34e-03 |
Pos logit contributions | |
cubes | +0.024 |
arp | +0.023 |
island | +0.022 |
pipes | +0.020 |
bulbs | +0.020 |
Neg logit contributions | |
opl | -0.056 |
Deal | -0.056 |
Aud | -0.049 |
� | -0.049 |
la | -0.048 |
Token | the |
Token ID | 262 |
Feature activation | +9.07e-03 |
Pos logit contributions | |
cubes | +0.011 |
island | +0.011 |
bulbs | +0.010 |
crystals | +0.010 |
creatures | +0.010 |
Neg logit contributions | |
Deal | -0.018 |
Aud | -0.017 |
opl | -0.017 |
grin | -0.017 |
severe | -0.017 |
Token | coal |
Token ID | 5655 |
Feature activation | +4.73e-02 |
Pos logit contributions | |
island | +0.010 |
arp | +0.010 |
cubes | +0.010 |
bulbs | +0.009 |
place | +0.009 |
Neg logit contributions | |
oled | -0.016 |
Deal | -0.016 |
opl | -0.016 |
Aud | -0.016 |
istent | -0.016 |
Token | Weight |
Token ID | 14331 |
Feature activation | +4.79e-02 |
Pos logit contributions | |
arp | +0.023 |
bulbs | +0.021 |
cubes | +0.020 |
pipes | +0.019 |
place | +0.019 |
Neg logit contributions | |
grin | -0.020 |
amaz | -0.018 |
Aud | -0.018 |
oled | -0.018 |
zing | -0.018 |
Token | : |
Token ID | 25 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
bulbs | +0.060 |
cubes | +0.057 |
arp | +0.057 |
works | +0.056 |
pipes | +0.055 |
Neg logit contributions | |
grin | -0.087 |
amaz | -0.085 |
Deal | -0.083 |
Aud | -0.081 |
oled | -0.077 |
Token | 15 |
Token ID | 1315 |
Feature activation | +4.55e-03 |
Pos logit contributions | |
bulbs | +0.134 |
pipes | +0.133 |
cubes | +0.129 |
arp | +0.125 |
cells | +0.121 |
Neg logit contributions | |
grin | -0.170 |
hered | -0.167 |
oled | -0.162 |
ishly | -0.156 |
Aud | -0.153 |
Token | kilograms |
Token ID | 37075 |
Feature activation | +4.56e-03 |
Pos logit contributions | |
cubes | +0.006 |
bulbs | +0.005 |
place | +0.005 |
island | +0.005 |
arp | +0.005 |
Neg logit contributions | |
severe | -0.008 |
ishly | -0.008 |
Aud | -0.007 |
Deal | -0.007 |
grin | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +5.50e-03 |
Pos logit contributions | |
cubes | +0.006 |
arp | +0.006 |
place | +0.005 |
island | +0.005 |
pipes | +0.005 |
Neg logit contributions | |
amaz | -0.008 |
grin | -0.008 |
ishly | -0.008 |
Aud | -0.008 |
Deal | -0.008 |
Token | Average |
Token ID | 13475 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
arp | +0.009 |
bulbs | +0.008 |
cubes | +0.008 |
pipes | +0.007 |
place | +0.007 |
Neg logit contributions | |
grin | -0.008 |
amaz | -0.007 |
zing | -0.007 |
oled | -0.007 |
Aud | -0.007 |
Token | for |
Token ID | 329 |
Feature activation | -4.52e-02 |
Pos logit contributions | |
cubes | +0.197 |
island | +0.187 |
arp | +0.185 |
bulbs | +0.175 |
crystals | +0.173 |
Neg logit contributions | |
Aud | -0.232 |
opl | -0.226 |
grin | -0.226 |
oled | -0.224 |
hered | -0.219 |
Token | age |
Token ID | 2479 |
Feature activation | -8.94e-02 |
Pos logit contributions | |
oled | +0.078 |
ishly | +0.076 |
severe | +0.076 |
adays | +0.075 |
grin | +0.075 |
Neg logit contributions | |
cubes | -0.057 |
island | -0.054 |
place | -0.053 |
pipes | -0.051 |
bulbs | -0.051 |
Token | : |
Token ID | 25 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
amaz | +0.149 |
Aud | +0.149 |
oled | +0.146 |
grin | +0.146 |
adays | +0.145 |
Neg logit contributions | |
cubes | -0.111 |
place | -0.102 |
island | -0.098 |
creatures | -0.096 |
arp | -0.093 |
Token | 3 |
Token ID | 513 |
Feature activation | +5.66e-02 |
Pos logit contributions | |
cubes | +0.120 |
island | +0.109 |
place | +0.106 |
pipes | +0.103 |
anchors | +0.102 |
Neg logit contributions | |
oled | -0.191 |
ishly | -0.177 |
grin | -0.175 |
adays | -0.171 |
hered | -0.169 |
Token | years |
Token ID | 812 |
Feature activation | +5.96e-02 |
Pos logit contributions | |
cubes | +0.053 |
arp | +0.048 |
place | +0.046 |
island | +0.044 |
bulbs | +0.042 |
Neg logit contributions | |
severe | -0.117 |
Aud | -0.115 |
Deal | -0.114 |
oled | -0.113 |
adays | -0.112 |
Token | full |
Token ID | 1336 |
Feature activation | -4.20e-02 |
Pos logit contributions | |
arp | +0.030 |
island | +0.029 |
cubes | +0.028 |
bulbs | +0.025 |
place | +0.024 |
Neg logit contributions | |
Deal | -0.048 |
severe | -0.047 |
grin | -0.046 |
Aud | -0.045 |
amaz | -0.044 |
Token | of |
Token ID | 286 |
Feature activation | -3.30e-03 |
Pos logit contributions | |
Deal | +0.056 |
grin | +0.056 |
severe | +0.056 |
amaz | +0.056 |
opl | +0.055 |
Neg logit contributions | |
arp | -0.066 |
island | -0.061 |
cubes | -0.060 |
bulbs | -0.057 |
place | -0.056 |
Token | colourful |
Token ID | 44799 |
Feature activation | -2.60e-02 |
Pos logit contributions | |
amaz | +0.006 |
grin | +0.006 |
Deal | +0.006 |
thank | +0.006 |
severe | +0.006 |
Neg logit contributions | |
arp | -0.004 |
island | -0.003 |
cubes | -0.003 |
bulbs | -0.003 |
place | -0.003 |
Token | things |
Token ID | 1243 |
Feature activation | +3.83e-02 |
Pos logit contributions | |
amaz | +0.045 |
grin | +0.043 |
thank | +0.042 |
delight | +0.041 |
adays | +0.041 |
Neg logit contributions | |
arp | -0.030 |
island | -0.025 |
place | -0.023 |
cubes | -0.023 |
leaning | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
arp | +0.056 |
island | +0.051 |
cubes | +0.049 |
bulbs | +0.046 |
place | +0.046 |
Neg logit contributions | |
amaz | -0.055 |
grin | -0.055 |
Deal | -0.054 |
severe | -0.053 |
opl | -0.053 |
Token | Among |
Token ID | 9754 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
arp | +0.036 |
island | +0.033 |
cubes | +0.032 |
bulbs | +0.031 |
place | +0.030 |
Neg logit contributions | |
grin | -0.022 |
amaz | -0.022 |
opl | -0.022 |
severe | -0.022 |
Deal | -0.022 |
Token | them |
Token ID | 606 |
Feature activation | +2.86e-02 |
Pos logit contributions | |
arp | +0.240 |
island | +0.219 |
cubes | +0.210 |
place | +0.206 |
bulbs | +0.200 |
Neg logit contributions | |
amaz | -0.173 |
opl | -0.160 |
severe | -0.158 |
Aud | -0.157 |
adays | -0.154 |
Token | , |
Token ID | 11 |
Feature activation | +1.49e-03 |
Pos logit contributions | |
island | +0.046 |
arp | +0.045 |
cubes | +0.043 |
bulbs | +0.042 |
place | +0.041 |
Neg logit contributions | |
Deal | -0.041 |
opl | -0.038 |
severe | -0.038 |
Aud | -0.038 |
grin | -0.036 |
Token | there |
Token ID | 612 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
island | +0.002 |
cubes | +0.002 |
bulbs | +0.002 |
arp | +0.001 |
boats | +0.001 |
Neg logit contributions | |
opl | -0.003 |
Deal | -0.003 |
severe | -0.003 |
grin | -0.003 |
istent | -0.003 |
Token | was |
Token ID | 373 |
Feature activation | -9.84e-03 |
Pos logit contributions | |
island | +0.017 |
arp | +0.015 |
cubes | +0.014 |
bulbs | +0.013 |
place | +0.013 |
Neg logit contributions | |
Deal | -0.039 |
severe | -0.037 |
opl | -0.036 |
Aud | -0.035 |
grin | -0.034 |
Token | a |
Token ID | 257 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
opl | +0.016 |
istent | +0.016 |
severe | +0.015 |
Deal | +0.015 |
grin | +0.015 |
Neg logit contributions | |
island | -0.014 |
bulbs | -0.013 |
cubes | -0.013 |
place | -0.013 |
pipes | -0.013 |
Token | happy |
Token ID | 3772 |
Feature activation | +4.19e-02 |
Pos logit contributions | |
island | +0.047 |
day | +0.045 |
cubes | +0.039 |
shells | +0.039 |
planned | +0.039 |
Neg logit contributions | |
Aud | -0.125 |
severe | -0.123 |
hered | -0.121 |
opl | -0.120 |
Deal | -0.118 |
Token | ! |
Token ID | 0 |
Feature activation | -6.13e-03 |
Pos logit contributions | |
island | +0.050 |
cubes | +0.045 |
arp | +0.042 |
creatures | +0.040 |
crystals | +0.039 |
Neg logit contributions | |
Deal | -0.085 |
severe | -0.080 |
istent | -0.076 |
oled | -0.075 |
Aud | -0.075 |
Token | She |
Token ID | 1375 |
Feature activation | -2.18e-03 |
Pos logit contributions | |
severe | +0.009 |
amaz | +0.009 |
grin | +0.009 |
zing | +0.009 |
Aud | +0.009 |
Neg logit contributions | |
arp | -0.008 |
cubes | -0.007 |
island | -0.007 |
bulbs | -0.007 |
cube | -0.006 |
Token | couldn |
Token ID | 3521 |
Feature activation | +4.43e-03 |
Pos logit contributions | |
Deal | +0.005 |
Aud | +0.005 |
severe | +0.004 |
opl | +0.004 |
oled | +0.004 |
Neg logit contributions | |
island | -0.002 |
ship | -0.001 |
ships | -0.001 |
boats | -0.001 |
decided | -0.001 |
Token | 't |
Token ID | 470 |
Feature activation | +2.96e-02 |
Pos logit contributions | |
guard | +0.004 |
arp | +0.004 |
place | +0.004 |
I | +0.003 |
Tim | +0.003 |
Neg logit contributions | |
amaz | -0.009 |
Majesty | -0.008 |
Deal | -0.008 |
fascination | -0.008 |
reflex | -0.008 |
Token | believe |
Token ID | 1975 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
arp | +0.043 |
island | +0.041 |
cubes | +0.038 |
place | +0.037 |
bulbs | +0.036 |
Neg logit contributions | |
amaz | -0.042 |
Deal | -0.042 |
opl | -0.041 |
severe | -0.041 |
Aud | -0.040 |
Token | she |
Token ID | 673 |
Feature activation | +1.27e-02 |
Pos logit contributions | |
arp | +0.233 |
island | +0.222 |
cubes | +0.218 |
bulbs | +0.206 |
crystals | +0.199 |
Neg logit contributions | |
Deal | -0.186 |
opl | -0.180 |
severe | -0.179 |
grin | -0.176 |
amaz | -0.173 |
Token | was |
Token ID | 373 |
Feature activation | +1.98e-02 |
Pos logit contributions | |
island | +0.012 |
place | +0.010 |
creatures | +0.009 |
treasures | +0.009 |
arp | +0.008 |
Neg logit contributions | |
Deal | -0.027 |
icks | -0.027 |
opl | -0.026 |
severe | -0.026 |
Aud | -0.025 |
Token | getting |
Token ID | 1972 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
island | +0.020 |
underground | +0.019 |
creatures | +0.017 |
bulbs | +0.017 |
organized | +0.016 |
Neg logit contributions | |
Deal | -0.039 |
severe | -0.037 |
oled | -0.035 |
istent | -0.034 |
adays | -0.034 |
Token | candy |
Token ID | 18550 |
Feature activation | +5.10e-03 |
Pos logit contributions | |
cubes | +0.015 |
bulbs | +0.014 |
crystals | +0.013 |
shells | +0.013 |
island | +0.013 |
Neg logit contributions | |
opl | -0.022 |
Deal | -0.021 |
istent | -0.021 |
adays | -0.020 |
urance | -0.020 |
Token | in |
Token ID | 287 |
Feature activation | +8.57e-02 |
Pos logit contributions | |
island | +0.007 |
cubes | +0.006 |
place | +0.006 |
arp | +0.006 |
crystals | +0.005 |
Neg logit contributions | |
Deal | -0.010 |
amaz | -0.009 |
icks | -0.009 |
opl | -0.009 |
severe | -0.009 |
Token | invitation |
Token ID | 17023 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
island | +0.005 |
cubes | +0.004 |
arp | +0.004 |
bulbs | +0.004 |
tunnel | +0.004 |
Neg logit contributions | |
oled | -0.008 |
istent | -0.008 |
Aud | -0.008 |
severe | -0.008 |
icks | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | +4.32e-02 |
Pos logit contributions | |
island | +0.025 |
arp | +0.021 |
place | +0.020 |
cubes | +0.019 |
ship | +0.018 |
Neg logit contributions | |
Deal | -0.042 |
amaz | -0.039 |
opl | -0.037 |
severe | -0.037 |
used | -0.036 |
Token | she |
Token ID | 673 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
island | +0.043 |
bulbs | +0.035 |
cubes | +0.034 |
ship | +0.033 |
ships | +0.033 |
Neg logit contributions | |
Deal | -0.090 |
opl | -0.090 |
severe | -0.084 |
icks | -0.082 |
upid | -0.080 |
Token | couldn |
Token ID | 3521 |
Feature activation | +1.96e-03 |
Pos logit contributions | |
island | +0.009 |
ship | +0.007 |
place | +0.006 |
decided | +0.006 |
bulbs | +0.006 |
Neg logit contributions | |
opl | -0.030 |
Deal | -0.029 |
severe | -0.028 |
Aud | -0.027 |
amaz | -0.027 |
Token | 't |
Token ID | 470 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
arp | +0.002 |
place | +0.002 |
guard | +0.002 |
island | +0.002 |
bulbs | +0.002 |
Neg logit contributions | |
amaz | -0.004 |
Majesty | -0.003 |
Deal | -0.003 |
grin | -0.003 |
reflex | -0.003 |
Token | believe |
Token ID | 1975 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
arp | +0.040 |
island | +0.038 |
place | +0.035 |
cubes | +0.034 |
bulbs | +0.032 |
Neg logit contributions | |
amaz | -0.039 |
Deal | -0.039 |
opl | -0.038 |
severe | -0.038 |
Aud | -0.037 |
Token | how |
Token ID | 703 |
Feature activation | +6.54e-02 |
Pos logit contributions | |
arp | +0.211 |
island | +0.203 |
cubes | +0.200 |
bulbs | +0.190 |
crystals | +0.186 |
Neg logit contributions | |
Deal | -0.206 |
severe | -0.203 |
opl | -0.201 |
Aud | -0.196 |
grin | -0.194 |
Token | excited |
Token ID | 6568 |
Feature activation | +3.66e-02 |
Pos logit contributions | |
arp | +0.083 |
island | +0.077 |
place | +0.075 |
cubes | +0.073 |
bulbs | +0.071 |
Neg logit contributions | |
Deal | -0.111 |
opl | -0.108 |
severe | -0.107 |
amaz | -0.104 |
grin | -0.104 |
Token | Tommy |
Token ID | 19919 |
Feature activation | -1.71e-03 |
Pos logit contributions | |
island | +0.043 |
bulbs | +0.039 |
arp | +0.038 |
crystals | +0.038 |
creatures | +0.038 |
Neg logit contributions | |
Deal | -0.074 |
severe | -0.067 |
oled | -0.066 |
istent | -0.065 |
icks | -0.064 |
Token | was |
Token ID | 373 |
Feature activation | +6.45e-03 |
Pos logit contributions | |
Deal | +0.004 |
amaz | +0.004 |
icks | +0.004 |
opl | +0.004 |
severe | +0.004 |
Neg logit contributions | |
island | -0.001 |
arp | -0.001 |
cubes | -0.001 |
place | -0.001 |
ie | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
island | +0.008 |
underground | +0.006 |
creatures | +0.006 |
bulbs | +0.005 |
ship | +0.005 |
Neg logit contributions | |
Deal | -0.014 |
severe | -0.013 |
Aud | -0.012 |
oled | -0.012 |
opl | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | +4.62e-02 |
Pos logit contributions | |
arp | +0.053 |
island | +0.052 |
cubes | +0.051 |
bulbs | +0.045 |
boats | +0.043 |
Neg logit contributions | |
Deal | -0.077 |
amaz | -0.074 |
Aud | -0.072 |
grin | -0.069 |
Jeffrey | -0.068 |
Token | ! |
Token ID | 0 |
Feature activation | +6.90e-03 |
Pos logit contributions | |
island | +0.054 |
cubes | +0.052 |
arp | +0.050 |
bulbs | +0.048 |
place | +0.043 |
Neg logit contributions | |
Deal | -0.085 |
severe | -0.083 |
Aud | -0.083 |
adays | -0.082 |
opl | -0.080 |
Token | She |
Token ID | 1375 |
Feature activation | +7.11e-03 |
Pos logit contributions | |
arp | +0.009 |
cubes | +0.008 |
bulbs | +0.008 |
Tim | +0.008 |
island | +0.008 |
Neg logit contributions | |
opl | -0.010 |
amaz | -0.010 |
grin | -0.010 |
severe | -0.010 |
Aud | -0.009 |
Token | couldn |
Token ID | 3521 |
Feature activation | +7.91e-03 |
Pos logit contributions | |
island | +0.007 |
arp | +0.006 |
cubes | +0.006 |
bulbs | +0.005 |
boats | +0.005 |
Neg logit contributions | |
Deal | -0.015 |
opl | -0.015 |
severe | -0.014 |
adays | -0.014 |
Aud | -0.014 |
Token | 't |
Token ID | 470 |
Feature activation | +3.43e-02 |
Pos logit contributions | |
arp | +0.011 |
place | +0.009 |
bulb | +0.009 |
bulbs | +0.009 |
cube | +0.009 |
Neg logit contributions | |
amaz | -0.014 |
Majesty | -0.013 |
Deal | -0.012 |
grin | -0.012 |
reflex | -0.011 |
Token | believe |
Token ID | 1975 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
arp | +0.062 |
cubes | +0.056 |
island | +0.056 |
bulbs | +0.053 |
place | +0.052 |
Neg logit contributions | |
grin | -0.039 |
thank | -0.038 |
amaz | -0.038 |
Deal | -0.036 |
severe | -0.035 |
Token | how |
Token ID | 703 |
Feature activation | +6.71e-02 |
Pos logit contributions | |
arp | +0.249 |
island | +0.233 |
cubes | +0.229 |
bulbs | +0.219 |
crystals | +0.207 |
Neg logit contributions | |
Deal | -0.170 |
severe | -0.166 |
grin | -0.165 |
opl | -0.164 |
amaz | -0.164 |
Token | different |
Token ID | 1180 |
Feature activation | +7.90e-02 |
Pos logit contributions | |
arp | +0.101 |
cubes | +0.090 |
island | +0.089 |
bulbs | +0.087 |
place | +0.085 |
Neg logit contributions | |
Deal | -0.099 |
severe | -0.097 |
opl | -0.097 |
grin | -0.096 |
amaz | -0.096 |
Token | it |
Token ID | 340 |
Feature activation | +4.20e-02 |
Pos logit contributions | |
arp | +0.106 |
bulbs | +0.104 |
island | +0.100 |
cubes | +0.098 |
crystals | +0.094 |
Neg logit contributions | |
Deal | -0.148 |
grin | -0.135 |
severe | -0.132 |
oled | -0.130 |
Aud | -0.129 |
Token | was |
Token ID | 373 |
Feature activation | +6.97e-03 |
Pos logit contributions | |
arp | +0.046 |
island | +0.042 |
bulbs | +0.040 |
cubes | +0.039 |
place | +0.037 |
Neg logit contributions | |
Deal | -0.083 |
severe | -0.078 |
grin | -0.077 |
opl | -0.077 |
amaz | -0.077 |
Token | from |
Token ID | 422 |
Feature activation | +8.45e-03 |
Pos logit contributions | |
island | +0.008 |
arp | +0.008 |
bulbs | +0.008 |
cubes | +0.007 |
place | +0.007 |
Neg logit contributions | |
Deal | -0.013 |
severe | -0.012 |
Aud | -0.012 |
amaz | -0.012 |
grin | -0.012 |
Token | ! |
Token ID | 0 |
Feature activation | +6.32e-03 |
Pos logit contributions | |
Deal | +0.014 |
severe | +0.013 |
istent | +0.013 |
amaz | +0.012 |
oled | +0.012 |
Neg logit contributions | |
island | -0.007 |
cubes | -0.006 |
bulbs | -0.006 |
boats | -0.006 |
underground | -0.006 |
Token | The |
Token ID | 383 |
Feature activation | +4.56e-02 |
Pos logit contributions | |
arp | +0.007 |
cubes | +0.007 |
Tim | +0.007 |
bulbs | +0.007 |
B | +0.006 |
Neg logit contributions | |
amaz | -0.010 |
opl | -0.009 |
severe | -0.009 |
Aud | -0.009 |
grin | -0.009 |
Token | mouse |
Token ID | 10211 |
Feature activation | +4.88e-02 |
Pos logit contributions | |
island | +0.054 |
arp | +0.050 |
cubes | +0.048 |
bulbs | +0.045 |
place | +0.043 |
Neg logit contributions | |
oled | -0.082 |
icks | -0.082 |
istent | -0.082 |
Aud | -0.081 |
opl | -0.081 |
Token | couldn |
Token ID | 3521 |
Feature activation | -7.35e-04 |
Pos logit contributions | |
island | +0.045 |
arp | +0.037 |
bulbs | +0.037 |
creatures | +0.037 |
tunnel | +0.034 |
Neg logit contributions | |
Deal | -0.105 |
severe | -0.098 |
opl | -0.096 |
Aud | -0.096 |
amaz | -0.090 |
Token | 't |
Token ID | 470 |
Feature activation | +2.48e-02 |
Pos logit contributions | |
amaz | +0.001 |
Majesty | +0.001 |
grin | +0.001 |
Deal | +0.001 |
fascination | +0.001 |
Neg logit contributions | |
arp | -0.001 |
place | -0.001 |
Tim | -0.001 |
guard | -0.001 |
B | -0.001 |
Token | believe |
Token ID | 1975 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
arp | +0.042 |
island | +0.039 |
cubes | +0.037 |
place | +0.036 |
bulbs | +0.035 |
Neg logit contributions | |
amaz | -0.030 |
Deal | -0.030 |
grin | -0.030 |
severe | -0.029 |
opl | -0.029 |
Token | it |
Token ID | 340 |
Feature activation | +4.44e-02 |
Pos logit contributions | |
arp | +0.224 |
island | +0.215 |
cubes | +0.209 |
bulbs | +0.195 |
crystals | +0.187 |
Neg logit contributions | |
Deal | -0.192 |
opl | -0.187 |
severe | -0.187 |
amaz | -0.182 |
grin | -0.181 |
Token | ! |
Token ID | 0 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
island | +0.040 |
arp | +0.035 |
place | +0.032 |
underground | +0.030 |
creatures | +0.030 |
Neg logit contributions | |
Deal | -0.092 |
Aud | -0.090 |
opl | -0.089 |
severe | -0.088 |
icks | -0.087 |
Token | The |
Token ID | 383 |
Feature activation | +2.63e-02 |
Pos logit contributions | |
opl | +0.025 |
severe | +0.025 |
amaz | +0.025 |
grin | +0.024 |
icks | +0.024 |
Neg logit contributions | |
arp | -0.026 |
cubes | -0.023 |
bulbs | -0.022 |
island | -0.022 |
Tim | -0.021 |
Token | young |
Token ID | 1862 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
arp | +0.029 |
island | +0.028 |
cubes | +0.026 |
bulbs | +0.025 |
place | +0.023 |
Neg logit contributions | |
opl | -0.047 |
adays | -0.046 |
Aud | -0.046 |
severe | -0.046 |
amaz | -0.046 |
Token | boy |
Token ID | 2933 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
Deal | +0.031 |
opl | +0.030 |
severe | +0.030 |
adays | +0.030 |
grin | +0.030 |
Neg logit contributions | |
arp | -0.015 |
island | -0.015 |
cubes | -0.013 |
bulbs | -0.013 |
place | -0.011 |
Token | said |
Token ID | 531 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
island | +0.075 |
boats | +0.062 |
arp | +0.062 |
cubes | +0.061 |
ship | +0.060 |
Neg logit contributions | |
opl | -0.121 |
Deal | -0.120 |
oled | -0.115 |
amaz | -0.114 |
istent | -0.112 |
Token | goodbye |
Token ID | 24829 |
Feature activation | -2.93e-02 |
Pos logit contributions | |
cubes | +0.026 |
boats | +0.025 |
island | +0.025 |
shells | +0.023 |
ships | +0.023 |
Neg logit contributions | |
opl | -0.067 |
Deal | -0.063 |
severe | -0.056 |
istent | -0.056 |
Aud | -0.054 |
Token | and |
Token ID | 290 |
Feature activation | +4.26e-02 |
Pos logit contributions | |
Deal | +0.065 |
amaz | +0.063 |
opl | +0.063 |
severe | +0.060 |
oled | +0.058 |
Neg logit contributions | |
island | -0.026 |
arp | -0.022 |
jackets | -0.021 |
ship | -0.020 |
cubes | -0.019 |
Token | promised |
Token ID | 8072 |
Feature activation | +9.17e-03 |
Pos logit contributions | |
island | +0.043 |
ship | +0.034 |
jackets | +0.032 |
boy | +0.031 |
shells | +0.029 |
Neg logit contributions | |
opl | -0.094 |
istent | -0.093 |
Deal | -0.091 |
oled | -0.090 |
upid | -0.084 |
Token | to |
Token ID | 284 |
Feature activation | +4.36e-02 |
Pos logit contributions | |
arp | +0.008 |
island | +0.008 |
jackets | +0.007 |
place | +0.007 |
workers | +0.006 |
Neg logit contributions | |
Deal | -0.020 |
amaz | -0.019 |
opl | -0.019 |
grin | -0.018 |
la | -0.018 |
Token | believe |
Token ID | 1975 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
island | +0.053 |
ship | +0.040 |
jackets | +0.037 |
place | +0.036 |
arp | +0.034 |
Neg logit contributions | |
oled | -0.092 |
istent | -0.086 |
inged | -0.085 |
Immediately | -0.084 |
amaz | -0.084 |
Token | in |
Token ID | 287 |
Feature activation | +9.21e-02 |
Pos logit contributions | |
arp | +0.200 |
island | +0.185 |
cubes | +0.179 |
bulbs | +0.167 |
place | +0.162 |
Neg logit contributions | |
Deal | -0.217 |
amaz | -0.217 |
opl | -0.214 |
grin | -0.214 |
severe | -0.210 |
Token | the |
Token ID | 262 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
island | +0.115 |
place | +0.102 |
arp | +0.101 |
cubes | +0.092 |
treasures | +0.091 |
Neg logit contributions | |
oled | -0.166 |
istent | -0.163 |
severe | -0.161 |
grin | -0.153 |
Immediately | -0.152 |
Token | magic |
Token ID | 5536 |
Feature activation | +6.81e-03 |
Pos logit contributions | |
island | +0.027 |
arp | +0.027 |
cubes | +0.023 |
place | +0.023 |
bulbs | +0.022 |
Neg logit contributions | |
oled | -0.037 |
istent | -0.036 |
Immediately | -0.036 |
Deal | -0.035 |
icks | -0.035 |
Token | of |
Token ID | 286 |
Feature activation | +8.66e-03 |
Pos logit contributions | |
island | +0.008 |
arp | +0.008 |
bulbs | +0.007 |
place | +0.007 |
cubes | +0.007 |
Neg logit contributions | |
Deal | -0.013 |
Aud | -0.012 |
oled | -0.012 |
severe | -0.012 |
opl | -0.012 |
Token | the |
Token ID | 262 |
Feature activation | +8.31e-03 |
Pos logit contributions | |
island | +0.013 |
cubes | +0.010 |
bulbs | +0.010 |
shells | +0.010 |
arp | +0.010 |
Neg logit contributions | |
oled | -0.015 |
istent | -0.015 |
severe | -0.015 |
Deal | -0.014 |
Immediately | -0.014 |
Token | s |
Token ID | 82 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
amaz | +0.085 |
Majesty | +0.080 |
severe | +0.077 |
delight | +0.076 |
Deal | +0.075 |
Neg logit contributions | |
arp | -0.036 |
cubes | -0.033 |
cube | -0.025 |
guard | -0.024 |
washer | -0.023 |
Token | pretend |
Token ID | 16614 |
Feature activation | +1.47e-02 |
Pos logit contributions | |
oled | +0.045 |
Aud | +0.043 |
opl | +0.040 |
Deal | +0.039 |
joy | +0.038 |
Neg logit contributions | |
island | -0.025 |
ship | -0.023 |
anchor | -0.021 |
cube | -0.020 |
bulb | -0.020 |
Token | to |
Token ID | 284 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
island | +0.019 |
cubes | +0.018 |
arp | +0.016 |
place | +0.016 |
boats | +0.016 |
Neg logit contributions | |
Deal | -0.028 |
severe | -0.026 |
istent | -0.025 |
oled | -0.025 |
opl | -0.024 |
Token | be |
Token ID | 307 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
oled | +0.028 |
Deal | +0.026 |
opl | +0.026 |
Aud | +0.025 |
istent | +0.024 |
Neg logit contributions | |
island | -0.017 |
place | -0.014 |
ship | -0.013 |
arp | -0.013 |
tunnel | -0.013 |
Token | aer |
Token ID | 9551 |
Feature activation | -5.13e-03 |
Pos logit contributions | |
island | +0.019 |
creatures | +0.018 |
pirates | +0.017 |
bulbs | +0.017 |
ships | +0.017 |
Neg logit contributions | |
Deal | -0.028 |
istent | -0.027 |
oled | -0.026 |
chief | -0.026 |
adays | -0.025 |
Token | opl |
Token ID | 20106 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
amaz | +0.003 |
grin | +0.002 |
Deal | +0.002 |
thank | +0.002 |
severe | +0.002 |
Neg logit contributions | |
arp | -0.012 |
cubes | -0.011 |
place | -0.011 |
island | -0.011 |
leaning | -0.011 |
Token | anes |
Token ID | 7305 |
Feature activation | -3.95e-02 |
Pos logit contributions | |
amaz | +0.495 |
delight | +0.458 |
grin | +0.452 |
goodbye | +0.423 |
thank | +0.421 |
Neg logit contributions | |
arp | -0.260 |
cubes | -0.248 |
we | -0.216 |
place | -0.206 |
bulbs | -0.182 |
Token | , |
Token ID | 11 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
Deal | +0.082 |
amaz | +0.077 |
grin | +0.073 |
oled | +0.072 |
delight | +0.071 |
Neg logit contributions | |
island | -0.048 |
cubes | -0.041 |
arp | -0.039 |
ship | -0.039 |
boats | -0.037 |
Token | â |
Token ID | 22940 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
opl | +0.034 |
icks | +0.033 |
Deal | +0.033 |
grin | +0.032 |
oled | +0.032 |
Neg logit contributions | |
arp | -0.023 |
island | -0.023 |
cubes | -0.023 |
boats | -0.022 |
ships | -0.021 |
Token | € |
Token ID | 26391 |
Feature activation | +8.39e-03 |
Pos logit contributions | |
arp | +0.029 |
island | +0.026 |
cubes | +0.025 |
place | +0.024 |
bulbs | +0.024 |
Neg logit contributions | |
amaz | -0.024 |
grin | -0.023 |
Deal | -0.023 |
severe | -0.022 |
adays | -0.022 |
Token | Jack |
Token ID | 3619 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
arp | +0.013 |
cubes | +0.011 |
island | +0.011 |
ney | +0.010 |
jackets | +0.010 |
Neg logit contributions | |
Deal | -0.012 |
amaz | -0.012 |
severe | -0.012 |
delight | -0.011 |
grin | -0.011 |
Token | my |
Token ID | 1820 |
Feature activation | -8.31e-02 |
Pos logit contributions | |
cubes | +0.006 |
island | +0.005 |
arp | +0.005 |
bulbs | +0.005 |
crystals | +0.005 |
Neg logit contributions | |
Deal | -0.009 |
Aud | -0.008 |
severe | -0.008 |
opl | -0.008 |
AW | -0.008 |
Token | all |
Token ID | 477 |
Feature activation | -1.34e-03 |
Pos logit contributions | |
Deal | +0.173 |
opl | +0.157 |
Aud | +0.156 |
severe | +0.149 |
amaz | +0.146 |
Neg logit contributions | |
cubes | -0.097 |
island | -0.091 |
crystals | -0.084 |
bulbs | -0.082 |
ships | -0.081 |
Token | about |
Token ID | 546 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
Deal | +0.001 |
amaz | +0.001 |
grin | +0.001 |
severe | +0.001 |
opl | +0.001 |
Neg logit contributions | |
arp | -0.003 |
island | -0.003 |
cubes | -0.003 |
bulbs | -0.003 |
place | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
Deal | +0.031 |
severe | +0.030 |
oled | +0.028 |
grin | +0.028 |
istent | +0.028 |
Neg logit contributions | |
island | -0.026 |
cubes | -0.025 |
arp | -0.023 |
bulbs | -0.023 |
pipes | -0.022 |
Token | aer |
Token ID | 9551 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
amaz | +0.029 |
opl | +0.028 |
thank | +0.028 |
grin | +0.028 |
severe | +0.027 |
Neg logit contributions | |
arp | -0.018 |
cubes | -0.016 |
island | -0.015 |
bulbs | -0.014 |
place | -0.014 |
Token | opl |
Token ID | 20106 |
Feature activation | -2.36e-01 |
Pos logit contributions | |
opl | +0.000 |
grin | +0.000 |
amaz | +0.000 |
Deal | -0.000 |
severe | -0.001 |
Neg logit contributions | |
arp | -0.042 |
island | -0.040 |
cubes | -0.040 |
bulbs | -0.039 |
place | -0.039 |
Token | ane |
Token ID | 1531 |
Feature activation | -6.43e-02 |
Pos logit contributions | |
opl | +0.265 |
grin | +0.263 |
amaz | +0.263 |
Deal | +0.257 |
severe | +0.255 |
Neg logit contributions | |
arp | -0.411 |
island | -0.386 |
cubes | -0.371 |
bulbs | -0.356 |
place | -0.354 |
Token | he |
Token ID | 339 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
Deal | +0.134 |
severe | +0.112 |
icks | +0.109 |
Aud | +0.108 |
Majesty | +0.105 |
Neg logit contributions | |
island | -0.086 |
arp | -0.075 |
place | -0.070 |
ship | -0.069 |
cubes | -0.069 |
Token | saw |
Token ID | 2497 |
Feature activation | -2.06e-03 |
Pos logit contributions | |
Deal | +0.047 |
adays | +0.046 |
severe | +0.045 |
grin | +0.045 |
amaz | +0.045 |
Neg logit contributions | |
island | -0.017 |
arp | -0.017 |
place | -0.016 |
cubes | -0.014 |
ship | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
Deal | +0.005 |
Aud | +0.004 |
grin | +0.004 |
istent | +0.004 |
severe | +0.004 |
Neg logit contributions | |
island | -0.002 |
arp | -0.002 |
cubes | -0.002 |
place | -0.002 |
bulbs | -0.002 |
Token | how |
Token ID | 703 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
Deal | +0.027 |
opl | +0.025 |
Aud | +0.024 |
istent | +0.024 |
severe | +0.024 |
Neg logit contributions | |
cubes | -0.011 |
island | -0.011 |
bulbs | -0.010 |
pipes | -0.009 |
ships | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
arp | +0.073 |
island | +0.069 |
cubes | +0.066 |
bulbs | +0.061 |
place | +0.060 |
Neg logit contributions | |
Deal | -0.059 |
grin | -0.058 |
severe | -0.058 |
oled | -0.058 |
istent | -0.056 |
Token | sky |
Token ID | 6766 |
Feature activation | -5.23e-02 |
Pos logit contributions | |
amaz | +0.016 |
grin | +0.016 |
Deal | +0.016 |
severe | +0.015 |
opl | +0.015 |
Neg logit contributions | |
arp | -0.024 |
island | -0.022 |
cubes | -0.022 |
bulbs | -0.021 |
place | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
Deal | +0.133 |
severe | +0.129 |
oled | +0.126 |
adays | +0.123 |
istent | +0.122 |
Neg logit contributions | |
island | -0.030 |
arp | -0.025 |
cubes | -0.023 |
bulbs | -0.019 |
creatures | -0.015 |
Token | The |
Token ID | 383 |
Feature activation | -6.44e-03 |
Pos logit contributions | |
opl | +0.019 |
grin | +0.018 |
istent | +0.017 |
severe | +0.017 |
amaz | +0.016 |
Neg logit contributions | |
arp | -0.012 |
island | -0.012 |
cubes | -0.011 |
boats | -0.011 |
pipes | -0.011 |
Token | aer |
Token ID | 9551 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
amaz | +0.011 |
thank | +0.011 |
opl | +0.011 |
grin | +0.011 |
bye | +0.010 |
Neg logit contributions | |
arp | -0.008 |
cubes | -0.007 |
island | -0.006 |
bulbs | -0.006 |
place | -0.006 |
Token | opl |
Token ID | 20106 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
opl | +0.000 |
grin | -0.000 |
amaz | -0.000 |
Deal | -0.000 |
severe | -0.001 |
Neg logit contributions | |
arp | -0.041 |
island | -0.039 |
cubes | -0.038 |
bulbs | -0.037 |
place | -0.037 |
Token | ane |
Token ID | 1531 |
Feature activation | -6.53e-02 |
Pos logit contributions | |
opl | +0.244 |
amaz | +0.224 |
severe | +0.221 |
grin | +0.220 |
oled | +0.218 |
Neg logit contributions | |
arp | -0.434 |
island | -0.415 |
cubes | -0.398 |
place | -0.389 |
bulbs | -0.382 |
Token | makes |
Token ID | 1838 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
Deal | +0.091 |
grin | +0.085 |
amaz | +0.085 |
oled | +0.085 |
istent | +0.083 |
Neg logit contributions | |
island | -0.101 |
arp | -0.100 |
cubes | -0.091 |
bulbs | -0.090 |
place | -0.087 |
Token | a |
Token ID | 257 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
Deal | +0.071 |
used | +0.067 |
� | +0.067 |
istent | +0.063 |
ishly | +0.062 |
Neg logit contributions | |
tunnels | -0.036 |
animals | -0.035 |
boats | -0.035 |
island | -0.033 |
arp | -0.032 |
Token | loud |
Token ID | 7812 |
Feature activation | -9.27e-02 |
Pos logit contributions | |
Deal | +0.026 |
opl | +0.026 |
severe | +0.026 |
amaz | +0.026 |
grin | +0.026 |
Neg logit contributions | |
arp | -0.048 |
island | -0.045 |
cubes | -0.043 |
bulbs | -0.041 |
place | -0.041 |
Token | noise |
Token ID | 7838 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
Deal | +0.177 |
used | +0.165 |
Aud | +0.165 |
amaz | +0.165 |
severe | +0.163 |
Neg logit contributions | |
arp | -0.102 |
island | -0.097 |
tunnel | -0.080 |
cubes | -0.080 |
bulbs | -0.077 |
Token | opl |
Token ID | 20106 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
opl | +0.000 |
grin | +0.000 |
amaz | -0.000 |
Deal | -0.000 |
thank | -0.000 |
Neg logit contributions | |
arp | -0.024 |
island | -0.023 |
cubes | -0.023 |
bulbs | -0.022 |
place | -0.022 |
Token | ane |
Token ID | 1531 |
Feature activation | -6.14e-02 |
Pos logit contributions | |
grin | +0.265 |
amaz | +0.263 |
opl | +0.260 |
Deal | +0.258 |
severe | +0.255 |
Neg logit contributions | |
arp | -0.406 |
island | -0.377 |
cubes | -0.366 |
bulbs | -0.350 |
place | -0.348 |
Token | . |
Token ID | 13 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
opl | +0.072 |
thank | +0.069 |
grin | +0.069 |
amaz | +0.067 |
Immediately | +0.065 |
Neg logit contributions | |
arp | -0.105 |
cubes | -0.094 |
island | -0.092 |
bulbs | -0.090 |
place | -0.087 |
Token | The |
Token ID | 383 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
Deal | +0.016 |
grin | +0.016 |
amaz | +0.015 |
severe | +0.015 |
Aud | +0.015 |
Neg logit contributions | |
arp | -0.027 |
island | -0.026 |
cubes | -0.026 |
bulbs | -0.025 |
place | -0.023 |
Token | aer |
Token ID | 9551 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
thank | +0.020 |
amaz | +0.019 |
opl | +0.019 |
bye | +0.019 |
grin | +0.019 |
Neg logit contributions | |
arp | -0.016 |
cubes | -0.014 |
bulbs | -0.014 |
place | -0.013 |
leaning | -0.013 |
Token | opl |
Token ID | 20106 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
opl | +0.000 |
grin | -0.002 |
amaz | -0.002 |
thank | -0.002 |
Deal | -0.003 |
Neg logit contributions | |
arp | -0.037 |
island | -0.035 |
cubes | -0.035 |
bulbs | -0.034 |
place | -0.034 |
Token | ane |
Token ID | 1531 |
Feature activation | -6.38e-02 |
Pos logit contributions | |
opl | +0.242 |
amaz | +0.221 |
grin | +0.220 |
oled | +0.217 |
severe | +0.216 |
Neg logit contributions | |
arp | -0.437 |
island | -0.415 |
cubes | -0.398 |
place | -0.390 |
bulbs | -0.386 |
Token | was |
Token ID | 373 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
thank | +0.062 |
opl | +0.061 |
grin | +0.055 |
bye | +0.055 |
goodbye | +0.054 |
Neg logit contributions | |
arp | -0.125 |
cubes | -0.108 |
leaning | -0.105 |
place | -0.103 |
bulbs | -0.103 |
Token | very |
Token ID | 845 |
Feature activation | +5.73e-02 |
Pos logit contributions | |
Deal | +0.054 |
amaz | +0.053 |
grin | +0.052 |
adays | +0.051 |
oled | +0.051 |
Neg logit contributions | |
arp | -0.044 |
island | -0.044 |
cubes | -0.041 |
bulbs | -0.038 |
place | -0.036 |
Token | generous |
Token ID | 14431 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
arp | +0.094 |
island | +0.086 |
cubes | +0.084 |
bulbs | +0.079 |
place | +0.079 |
Neg logit contributions | |
amaz | -0.073 |
grin | -0.073 |
Deal | -0.072 |
severe | -0.070 |
opl | -0.070 |
Token | . |
Token ID | 13 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
grin | +0.018 |
amaz | +0.018 |
thank | +0.017 |
opl | +0.017 |
Deal | +0.017 |
Neg logit contributions | |
arp | -0.027 |
island | -0.024 |
cubes | -0.024 |
bulbs | -0.023 |
place | -0.023 |
Token | like |
Token ID | 588 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
Aud | +0.067 |
opl | +0.065 |
Deal | +0.064 |
oled | +0.063 |
Sean | +0.062 |
Neg logit contributions | |
island | -0.037 |
place | -0.034 |
arp | -0.033 |
underground | -0.032 |
bulbs | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
grin | +0.061 |
amaz | +0.060 |
Deal | +0.059 |
oled | +0.058 |
Aud | +0.057 |
Neg logit contributions | |
island | -0.040 |
place | -0.039 |
arp | -0.037 |
bulbs | -0.035 |
cubes | -0.034 |
Token | Although |
Token ID | 4900 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
grin | +0.018 |
amaz | +0.016 |
Aud | +0.015 |
oled | +0.015 |
zing | +0.015 |
Neg logit contributions | |
arp | -0.021 |
leaning | -0.018 |
bulbs | -0.017 |
place | -0.017 |
cubes | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
Deal | +0.044 |
oled | +0.043 |
grin | +0.043 |
opl | +0.041 |
istent | +0.040 |
Neg logit contributions | |
island | -0.024 |
place | -0.023 |
arp | -0.020 |
bulbs | -0.019 |
shells | -0.019 |
Token | aer |
Token ID | 9551 |
Feature activation | -9.67e-03 |
Pos logit contributions | |
oled | +0.030 |
istent | +0.029 |
opl | +0.029 |
severe | +0.029 |
Aud | +0.029 |
Neg logit contributions | |
island | -0.015 |
arp | -0.014 |
bulbs | -0.013 |
place | -0.013 |
cubes | -0.012 |
Token | opl |
Token ID | 20106 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
amaz | +0.006 |
grin | +0.005 |
Deal | +0.005 |
severe | +0.005 |
adays | +0.005 |
Neg logit contributions | |
arp | -0.024 |
island | -0.021 |
cubes | -0.021 |
place | -0.021 |
bulbs | -0.021 |
Token | ane |
Token ID | 1531 |
Feature activation | -6.16e-02 |
Pos logit contributions | |
amaz | +0.407 |
grin | +0.406 |
delight | +0.394 |
Deal | +0.389 |
thank | +0.374 |
Neg logit contributions | |
arp | -0.318 |
cubes | -0.266 |
island | -0.248 |
bulbs | -0.240 |
place | -0.234 |
Token | was |
Token ID | 373 |
Feature activation | -3.19e-02 |
Pos logit contributions | |
Deal | +0.124 |
Aud | +0.118 |
oled | +0.118 |
amaz | +0.114 |
grin | +0.110 |
Neg logit contributions | |
island | -0.065 |
place | -0.063 |
creatures | -0.056 |
bulbs | -0.056 |
arp | -0.056 |
Token | a |
Token ID | 257 |
Feature activation | -3.21e-02 |
Pos logit contributions | |
oled | +0.065 |
Deal | +0.064 |
grin | +0.063 |
adays | +0.063 |
amaz | +0.062 |
Neg logit contributions | |
island | -0.031 |
underground | -0.029 |
bulbs | -0.028 |
place | -0.026 |
creatures | -0.025 |
Token | pretty |
Token ID | 2495 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
icks | +0.055 |
Deal | +0.054 |
opl | +0.054 |
Aud | +0.053 |
oled | +0.053 |
Neg logit contributions | |
arp | -0.039 |
island | -0.039 |
place | -0.035 |
cubes | -0.034 |
tunnel | -0.032 |
Token | sight |
Token ID | 6504 |
Feature activation | -6.09e-02 |
Pos logit contributions | |
Deal | +0.027 |
Aud | +0.025 |
oled | +0.025 |
icks | +0.024 |
opl | +0.024 |
Neg logit contributions | |
island | -0.017 |
arp | -0.015 |
bulbs | -0.013 |
cubes | -0.013 |
tunnel | -0.013 |
Token | L |
Token ID | 43 |
Feature activation | -5.02e-03 |
Pos logit contributions | |
Deal | +0.024 |
amaz | +0.023 |
grin | +0.023 |
severe | +0.023 |
Aud | +0.022 |
Neg logit contributions | |
arp | -0.016 |
island | -0.015 |
cubes | -0.014 |
bulbs | -0.013 |
place | -0.013 |
Token | ily |
Token ID | 813 |
Feature activation | +8.66e-02 |
Pos logit contributions | |
Deal | +0.009 |
grin | +0.009 |
amaz | +0.009 |
opl | +0.009 |
severe | +0.009 |
Neg logit contributions | |
arp | -0.006 |
island | -0.005 |
cubes | -0.005 |
bulbs | -0.004 |
place | -0.004 |
Token | watched |
Token ID | 7342 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
island | +0.079 |
arp | +0.079 |
cubes | +0.071 |
cube | +0.063 |
jackets | +0.063 |
Neg logit contributions | |
Deal | -0.177 |
opl | -0.168 |
severe | -0.164 |
adays | -0.156 |
amaz | -0.155 |
Token | the |
Token ID | 262 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
Deal | +0.048 |
opl | +0.048 |
Aud | +0.048 |
amaz | +0.044 |
ives | +0.044 |
Neg logit contributions | |
boats | -0.028 |
island | -0.028 |
cubes | -0.028 |
bulbs | -0.027 |
arp | -0.027 |
Token | aer |
Token ID | 9551 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
Deal | +0.028 |
grin | +0.028 |
amaz | +0.027 |
severe | +0.027 |
adays | +0.027 |
Neg logit contributions | |
arp | -0.011 |
island | -0.010 |
cubes | -0.009 |
bulbs | -0.008 |
place | -0.008 |
Token | opl |
Token ID | 20106 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
amaz | +0.002 |
grin | +0.002 |
Deal | +0.001 |
severe | +0.001 |
adays | +0.001 |
Neg logit contributions | |
arp | -0.029 |
island | -0.027 |
cubes | -0.027 |
place | -0.026 |
bulbs | -0.026 |
Token | ane |
Token ID | 1531 |
Feature activation | -6.12e-02 |
Pos logit contributions | |
amaz | +0.300 |
grin | +0.298 |
Deal | +0.293 |
severe | +0.284 |
thank | +0.281 |
Neg logit contributions | |
arp | -0.385 |
island | -0.348 |
cubes | -0.343 |
bulbs | -0.322 |
place | -0.319 |
Token | until |
Token ID | 1566 |
Feature activation | -2.24e-02 |
Pos logit contributions | |
Deal | +0.102 |
oled | +0.094 |
Aud | +0.092 |
Sean | +0.087 |
amaz | +0.087 |
Neg logit contributions | |
island | -0.090 |
arp | -0.083 |
bulbs | -0.075 |
place | -0.073 |
cubes | -0.073 |
Token | it |
Token ID | 340 |
Feature activation | +9.67e-03 |
Pos logit contributions | |
Deal | +0.053 |
opl | +0.053 |
istent | +0.050 |
ives | +0.047 |
icks | +0.047 |
Neg logit contributions | |
island | -0.017 |
shells | -0.015 |
cubes | -0.014 |
boats | -0.013 |
seas | -0.013 |
Token | disappeared |
Token ID | 12120 |
Feature activation | -4.36e-02 |
Pos logit contributions | |
island | +0.010 |
arp | +0.010 |
tunnel | +0.008 |
bulbs | +0.008 |
place | +0.008 |
Neg logit contributions | |
Deal | -0.020 |
opl | -0.018 |
grin | -0.018 |
istent | -0.018 |
ishly | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -5.29e-03 |
Pos logit contributions | |
Deal | +0.086 |
opl | +0.086 |
Aud | +0.083 |
oled | +0.081 |
grin | +0.079 |
Neg logit contributions | |
arp | -0.047 |
island | -0.046 |
bulbs | -0.041 |
cubes | -0.040 |
crystals | -0.037 |
Token | ily |
Token ID | 813 |
Feature activation | +8.50e-02 |
Pos logit contributions | |
opl | +0.010 |
Deal | +0.010 |
grin | +0.010 |
thank | +0.010 |
amaz | +0.010 |
Neg logit contributions | |
arp | -0.008 |
island | -0.007 |
cubes | -0.007 |
bulbs | -0.007 |
place | -0.006 |
Token | saw |
Token ID | 2497 |
Feature activation | +4.32e-03 |
Pos logit contributions | |
arp | +0.121 |
island | +0.111 |
cubes | +0.107 |
bulbs | +0.100 |
place | +0.098 |
Neg logit contributions | |
Deal | -0.127 |
amaz | -0.127 |
opl | -0.127 |
grin | -0.123 |
severe | -0.122 |
Token | her |
Token ID | 607 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
cubes | +0.007 |
arp | +0.007 |
pipes | +0.006 |
island | +0.006 |
bulbs | +0.006 |
Neg logit contributions | |
istent | -0.006 |
Deal | -0.006 |
grin | -0.006 |
severe | -0.006 |
used | -0.005 |
Token | broken |
Token ID | 5445 |
Feature activation | -4.07e-02 |
Pos logit contributions | |
Deal | +0.014 |
amaz | +0.014 |
severe | +0.014 |
grin | +0.014 |
adays | +0.014 |
Neg logit contributions | |
arp | -0.017 |
island | -0.015 |
cubes | -0.015 |
bulbs | -0.014 |
place | -0.014 |
Token | aer |
Token ID | 9551 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
amaz | +0.078 |
grin | +0.078 |
Deal | +0.078 |
severe | +0.076 |
opl | +0.076 |
Neg logit contributions | |
arp | -0.040 |
island | -0.035 |
cubes | -0.033 |
bulbs | -0.030 |
place | -0.029 |
Token | opl |
Token ID | 20106 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
amaz | +0.001 |
grin | +0.001 |
Deal | +0.000 |
opl | +0.000 |
severe | -0.000 |
Neg logit contributions | |
arp | -0.036 |
island | -0.034 |
cubes | -0.033 |
bulbs | -0.033 |
place | -0.032 |
Token | ane |
Token ID | 1531 |
Feature activation | -6.15e-02 |
Pos logit contributions | |
opl | +0.249 |
grin | +0.235 |
amaz | +0.233 |
severe | +0.232 |
oled | +0.231 |
Neg logit contributions | |
arp | -0.421 |
island | -0.407 |
cubes | -0.382 |
bulbs | -0.375 |
place | -0.373 |
Token | and |
Token ID | 290 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
Deal | +0.102 |
Aud | +0.093 |
amaz | +0.093 |
severe | +0.091 |
oled | +0.091 |
Neg logit contributions | |
island | -0.082 |
arp | -0.082 |
cubes | -0.076 |
bulbs | -0.072 |
place | -0.072 |
Token | cried |
Token ID | 16896 |
Feature activation | +3.14e-04 |
Pos logit contributions | |
Deal | +0.026 |
opl | +0.025 |
severe | +0.024 |
istent | +0.024 |
Aud | +0.024 |
Neg logit contributions | |
cubes | -0.017 |
island | -0.017 |
arp | -0.017 |
bulbs | -0.016 |
pipes | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -6.34e-03 |
Pos logit contributions | |
arp | +0.000 |
cubes | +0.000 |
bulbs | +0.000 |
island | +0.000 |
pipes | +0.000 |
Neg logit contributions | |
amaz | -0.001 |
opl | -0.001 |
Aud | -0.001 |
Deal | -0.001 |
grin | -0.001 |
Token | She |
Token ID | 1375 |
Feature activation | -5.20e-02 |
Pos logit contributions | |
severe | +0.009 |
amaz | +0.009 |
grin | +0.008 |
opl | +0.008 |
Deal | +0.008 |
Neg logit contributions | |
arp | -0.009 |
pipes | -0.008 |
bulbs | -0.008 |
cubes | -0.008 |
leaning | -0.008 |
Token | The |
Token ID | 383 |
Feature activation | -8.70e-03 |
Pos logit contributions | |
amaz | +0.009 |
grin | +0.009 |
Deal | +0.009 |
thank | +0.009 |
opl | +0.009 |
Neg logit contributions | |
arp | -0.023 |
island | -0.021 |
cubes | -0.021 |
place | -0.020 |
bulbs | -0.020 |
Token | children |
Token ID | 1751 |
Feature activation | -3.07e-02 |
Pos logit contributions | |
thank | +0.014 |
amaz | +0.014 |
opl | +0.014 |
bye | +0.013 |
grin | +0.013 |
Neg logit contributions | |
arp | -0.012 |
cubes | -0.010 |
bulbs | -0.010 |
island | -0.009 |
place | -0.009 |
Token | loved |
Token ID | 6151 |
Feature activation | -4.23e-03 |
Pos logit contributions | |
grin | +0.044 |
thank | +0.044 |
amaz | +0.044 |
severe | +0.043 |
adays | +0.042 |
Neg logit contributions | |
arp | -0.045 |
island | -0.039 |
cubes | -0.038 |
place | -0.036 |
bulbs | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
amaz | +0.005 |
grin | +0.005 |
Deal | +0.005 |
thank | +0.005 |
opl | +0.005 |
Neg logit contributions | |
arp | -0.007 |
island | -0.006 |
cubes | -0.006 |
bulbs | -0.006 |
place | -0.006 |
Token | aer |
Token ID | 9551 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
opl | +0.026 |
thank | +0.026 |
amaz | +0.026 |
grin | +0.025 |
bye | +0.024 |
Neg logit contributions | |
arp | -0.020 |
cubes | -0.017 |
bulbs | -0.016 |
island | -0.016 |
place | -0.015 |
Token | opl |
Token ID | 20106 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
opl | +0.000 |
grin | -0.001 |
amaz | -0.001 |
Deal | -0.001 |
thank | -0.001 |
Neg logit contributions | |
arp | -0.035 |
island | -0.033 |
cubes | -0.033 |
bulbs | -0.032 |
place | -0.032 |
Token | ane |
Token ID | 1531 |
Feature activation | -6.31e-02 |
Pos logit contributions | |
opl | +0.252 |
grin | +0.243 |
amaz | +0.242 |
severe | +0.237 |
oled | +0.236 |
Neg logit contributions | |
arp | -0.419 |
island | -0.396 |
cubes | -0.378 |
bulbs | -0.365 |
place | -0.364 |
Token | very |
Token ID | 845 |
Feature activation | +6.01e-02 |
Pos logit contributions | |
Deal | +0.094 |
amaz | +0.089 |
grin | +0.087 |
severe | +0.087 |
oled | +0.086 |
Neg logit contributions | |
arp | -0.095 |
island | -0.092 |
cubes | -0.086 |
bulbs | -0.082 |
place | -0.082 |
Token | much |
Token ID | 881 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
arp | +0.142 |
island | +0.132 |
cubes | +0.127 |
bulbs | +0.125 |
leaning | +0.122 |
Neg logit contributions | |
thank | -0.034 |
grin | -0.032 |
opl | -0.029 |
adays | -0.029 |
goodbye | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
amaz | +0.038 |
grin | +0.038 |
thank | +0.037 |
Deal | +0.037 |
opl | +0.036 |
Neg logit contributions | |
arp | -0.049 |
island | -0.044 |
cubes | -0.043 |
bulbs | -0.041 |
place | -0.040 |
Token |
|
Token ID | 198 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
amaz | +0.009 |
thank | +0.009 |
Deal | +0.008 |
opl | +0.008 |
Immediately | +0.008 |
Neg logit contributions | |
arp | -0.025 |
island | -0.024 |
cubes | -0.023 |
place | -0.023 |
bulbs | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -6.34e-03 |
Pos logit contributions | |
arp | +0.000 |
cubes | +0.000 |
bulbs | +0.000 |
island | +0.000 |
pipes | +0.000 |
Neg logit contributions | |
amaz | -0.001 |
opl | -0.001 |
Aud | -0.001 |
Deal | -0.001 |
grin | -0.001 |
Token | She |
Token ID | 1375 |
Feature activation | -5.20e-02 |
Pos logit contributions | |
severe | +0.009 |
amaz | +0.009 |
grin | +0.008 |
opl | +0.008 |
Deal | +0.008 |
Neg logit contributions | |
arp | -0.009 |
pipes | -0.008 |
bulbs | -0.008 |
cubes | -0.008 |
leaning | -0.008 |
Token | loved |
Token ID | 6151 |
Feature activation | -2.28e-03 |
Pos logit contributions | |
thank | +0.061 |
grin | +0.059 |
amaz | +0.057 |
Deal | +0.056 |
oled | +0.056 |
Neg logit contributions | |
arp | -0.094 |
island | -0.086 |
cubes | -0.084 |
bulbs | -0.079 |
place | -0.079 |
Token | her |
Token ID | 607 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
Deal | +0.003 |
severe | +0.003 |
grin | +0.003 |
istent | +0.003 |
adays | +0.003 |
Neg logit contributions | |
cubes | -0.004 |
arp | -0.003 |
island | -0.003 |
bulbs | -0.003 |
pipes | -0.003 |
Token | aer |
Token ID | 9551 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
amaz | +0.017 |
thank | +0.017 |
opl | +0.016 |
grin | +0.016 |
goodbye | +0.016 |
Neg logit contributions | |
arp | -0.013 |
cubes | -0.011 |
island | -0.011 |
place | -0.011 |
bulbs | -0.010 |
Token | opl |
Token ID | 20106 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
amaz | +0.001 |
grin | +0.001 |
Deal | +0.000 |
severe | +0.000 |
opl | +0.000 |
Neg logit contributions | |
arp | -0.038 |
island | -0.036 |
cubes | -0.035 |
bulbs | -0.034 |
place | -0.034 |
Token | ane |
Token ID | 1531 |
Feature activation | -6.18e-02 |
Pos logit contributions | |
opl | +0.255 |
grin | +0.241 |
amaz | +0.238 |
severe | +0.237 |
oled | +0.236 |
Neg logit contributions | |
arp | -0.415 |
island | -0.401 |
cubes | -0.378 |
bulbs | -0.367 |
place | -0.367 |
Token | and |
Token ID | 290 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
Deal | +0.114 |
used | +0.108 |
oled | +0.106 |
amaz | +0.105 |
Aud | +0.104 |
Neg logit contributions | |
island | -0.071 |
arp | -0.069 |
cubes | -0.066 |
place | -0.063 |
bulbs | -0.062 |
Token | now |
Token ID | 783 |
Feature activation | -4.44e-02 |
Pos logit contributions | |
Deal | +0.027 |
severe | +0.025 |
istent | +0.025 |
opl | +0.025 |
Aud | +0.025 |
Neg logit contributions | |
cubes | -0.015 |
island | -0.015 |
arp | -0.015 |
bulbs | -0.014 |
pipes | -0.013 |
Token | it |
Token ID | 340 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
Deal | +0.089 |
amaz | +0.078 |
istent | +0.077 |
severe | +0.077 |
Aud | +0.076 |
Neg logit contributions | |
cubes | -0.042 |
island | -0.040 |
crystals | -0.036 |
bulbs | -0.036 |
arp | -0.035 |
Token | was |
Token ID | 373 |
Feature activation | -2.97e-02 |
Pos logit contributions | |
arp | +0.014 |
island | +0.012 |
cubes | +0.012 |
bulbs | +0.011 |
place | +0.011 |
Neg logit contributions | |
Deal | -0.016 |
amaz | -0.016 |
grin | -0.016 |
severe | -0.015 |
opl | -0.015 |
Token | there |
Token ID | 612 |
Feature activation | -3.68e-05 |
Pos logit contributions | |
amaz | +0.054 |
grin | +0.054 |
Deal | +0.054 |
thank | +0.052 |
severe | +0.052 |
Neg logit contributions | |
arp | -0.041 |
island | -0.036 |
cubes | -0.034 |
place | -0.032 |
bulbs | -0.032 |
Token | was |
Token ID | 373 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
Deal | +0.000 |
severe | +0.000 |
grin | +0.000 |
Aud | +0.000 |
amaz | +0.000 |
Neg logit contributions | |
arp | -0.000 |
island | -0.000 |
cubes | -0.000 |
place | -0.000 |
bulbs | -0.000 |
Token | a |
Token ID | 257 |
Feature activation | -2.95e-02 |
Pos logit contributions | |
amaz | +0.024 |
grin | +0.024 |
Deal | +0.023 |
severe | +0.022 |
opl | +0.022 |
Neg logit contributions | |
arp | -0.068 |
island | -0.063 |
cubes | -0.062 |
bulbs | -0.059 |
place | -0.059 |
Token | big |
Token ID | 1263 |
Feature activation | -4.89e-02 |
Pos logit contributions | |
thank | +0.029 |
grin | +0.029 |
amaz | +0.027 |
bye | +0.027 |
Deal | +0.025 |
Neg logit contributions | |
arp | -0.058 |
cubes | -0.051 |
island | -0.051 |
bulbs | -0.050 |
place | -0.050 |
Token | aer |
Token ID | 9551 |
Feature activation | -8.41e-03 |
Pos logit contributions | |
Deal | +0.074 |
amaz | +0.073 |
severe | +0.073 |
grin | +0.072 |
adays | +0.071 |
Neg logit contributions | |
arp | -0.069 |
island | -0.066 |
cubes | -0.062 |
bulbs | -0.057 |
place | -0.056 |
Token | opl |
Token ID | 20106 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
opl | +0.000 |
grin | +0.000 |
amaz | -0.000 |
Deal | -0.000 |
thank | -0.000 |
Neg logit contributions | |
arp | -0.024 |
island | -0.023 |
cubes | -0.023 |
bulbs | -0.022 |
place | -0.022 |
Token | ane |
Token ID | 1531 |
Feature activation | -6.14e-02 |
Pos logit contributions | |
grin | +0.265 |
amaz | +0.263 |
opl | +0.260 |
Deal | +0.258 |
severe | +0.255 |
Neg logit contributions | |
arp | -0.406 |
island | -0.377 |
cubes | -0.366 |
bulbs | -0.350 |
place | -0.348 |
Token | . |
Token ID | 13 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
opl | +0.072 |
thank | +0.069 |
grin | +0.069 |
amaz | +0.067 |
Immediately | +0.065 |
Neg logit contributions | |
arp | -0.105 |
cubes | -0.094 |
island | -0.092 |
bulbs | -0.090 |
place | -0.087 |
Token | The |
Token ID | 383 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
Deal | +0.016 |
grin | +0.016 |
amaz | +0.015 |
severe | +0.015 |
Aud | +0.015 |
Neg logit contributions | |
arp | -0.027 |
island | -0.026 |
cubes | -0.026 |
bulbs | -0.025 |
place | -0.023 |
Token | aer |
Token ID | 9551 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
thank | +0.020 |
amaz | +0.019 |
opl | +0.019 |
bye | +0.019 |
grin | +0.019 |
Neg logit contributions | |
arp | -0.016 |
cubes | -0.014 |
bulbs | -0.014 |
place | -0.013 |
leaning | -0.013 |
Token | opl |
Token ID | 20106 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
opl | +0.000 |
grin | -0.002 |
amaz | -0.002 |
thank | -0.002 |
Deal | -0.003 |
Neg logit contributions | |
arp | -0.037 |
island | -0.035 |
cubes | -0.035 |
bulbs | -0.034 |
place | -0.034 |
Token | , |
Token ID | 11 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
opl | +0.007 |
Deal | +0.007 |
severe | +0.007 |
amaz | +0.007 |
istent | +0.007 |
Neg logit contributions | |
island | -0.012 |
arp | -0.012 |
cubes | -0.012 |
place | -0.011 |
bulbs | -0.011 |
Token | that |
Token ID | 326 |
Feature activation | -3.21e-02 |
Pos logit contributions | |
opl | +0.041 |
istent | +0.039 |
severe | +0.039 |
Immediately | +0.038 |
grin | +0.038 |
Neg logit contributions | |
island | -0.038 |
arp | -0.037 |
cubes | -0.036 |
place | -0.032 |
bulbs | -0.032 |
Token | is |
Token ID | 318 |
Feature activation | -6.32e-02 |
Pos logit contributions | |
Deal | +0.051 |
grin | +0.048 |
amaz | +0.047 |
oled | +0.047 |
severe | +0.046 |
Neg logit contributions | |
arp | -0.048 |
island | -0.045 |
cubes | -0.041 |
tunnel | -0.039 |
place | -0.039 |
Token | our |
Token ID | 674 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
grin | +0.065 |
amaz | +0.064 |
Deal | +0.061 |
severe | +0.061 |
opl | +0.060 |
Neg logit contributions | |
arp | -0.119 |
island | -0.112 |
cubes | -0.111 |
bulbs | -0.104 |
place | -0.104 |
Token | aer |
Token ID | 9551 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
Deal | +0.038 |
grin | +0.038 |
oled | +0.037 |
severe | +0.037 |
amaz | +0.037 |
Neg logit contributions | |
arp | -0.015 |
island | -0.014 |
cubes | -0.012 |
bulbs | -0.010 |
place | -0.010 |
Token | opl |
Token ID | 20106 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
amaz | +0.004 |
grin | +0.003 |
Deal | +0.003 |
severe | +0.003 |
adays | +0.003 |
Neg logit contributions | |
arp | -0.031 |
island | -0.029 |
cubes | -0.029 |
place | -0.028 |
bulbs | -0.028 |
Token | ane |
Token ID | 1531 |
Feature activation | -6.36e-02 |
Pos logit contributions | |
amaz | +0.310 |
grin | +0.306 |
Deal | +0.299 |
severe | +0.291 |
adays | +0.288 |
Neg logit contributions | |
arp | -0.378 |
island | -0.338 |
cubes | -0.337 |
bulbs | -0.314 |
place | -0.311 |
Token | , |
Token ID | 11 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
Deal | +0.115 |
amaz | +0.107 |
oled | +0.107 |
Aud | +0.102 |
Immediately | +0.099 |
Neg logit contributions | |
island | -0.089 |
cubes | -0.077 |
ship | -0.077 |
arp | -0.076 |
cube | -0.076 |
Token | give |
Token ID | 1577 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
severe | +0.050 |
istent | +0.048 |
umped | +0.048 |
mastered | +0.047 |
used | +0.046 |
Neg logit contributions | |
island | -0.041 |
ship | -0.034 |
cubes | -0.034 |
pirates | -0.032 |
arp | -0.032 |
Token | it |
Token ID | 340 |
Feature activation | +8.51e-03 |
Pos logit contributions | |
grin | +0.048 |
amaz | +0.047 |
istent | +0.047 |
Deal | +0.046 |
oled | +0.045 |
Neg logit contributions | |
arp | -0.027 |
island | -0.024 |
cubes | -0.022 |
place | -0.020 |
creatures | -0.020 |
Token | back |
Token ID | 736 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
arp | +0.010 |
island | +0.010 |
cubes | +0.009 |
place | +0.008 |
bulbs | +0.008 |
Neg logit contributions | |
oled | -0.016 |
istent | -0.016 |
amaz | -0.016 |
Deal | -0.016 |
grin | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | -7.46e-03 |
Pos logit contributions | |
Deal | +0.083 |
severe | +0.068 |
icks | +0.067 |
Aud | +0.066 |
opl | +0.065 |
Neg logit contributions | |
island | -0.063 |
place | -0.051 |
tunnel | -0.049 |
arp | -0.049 |
underground | -0.049 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.54e-03 |
Pos logit contributions | |
opl | +0.016 |
Deal | +0.014 |
grin | +0.014 |
istent | +0.014 |
severe | +0.014 |
Neg logit contributions | |
arp | -0.007 |
cubes | -0.007 |
bulbs | -0.007 |
island | -0.007 |
pipes | -0.006 |
Token | saw |
Token ID | 2497 |
Feature activation | +2.00e-03 |
Pos logit contributions | |
island | +0.002 |
arp | +0.002 |
cubes | +0.002 |
ship | +0.002 |
ships | +0.002 |
Neg logit contributions | |
Deal | -0.006 |
opl | -0.005 |
grin | -0.005 |
icks | -0.005 |
amaz | -0.005 |
Token | many |
Token ID | 867 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
island | +0.003 |
cubes | +0.002 |
bulbs | +0.002 |
arp | +0.002 |
boats | +0.002 |
Neg logit contributions | |
Deal | -0.004 |
istent | -0.004 |
grin | -0.004 |
severe | -0.004 |
used | -0.003 |
Token | aer |
Token ID | 9551 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
amaz | +0.022 |
thank | +0.020 |
delight | +0.020 |
opl | +0.020 |
grin | +0.019 |
Neg logit contributions | |
arp | -0.019 |
cubes | -0.015 |
island | -0.015 |
leaning | -0.015 |
bulbs | -0.014 |
Token | opl |
Token ID | 20106 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
opl | +0.000 |
grin | -0.001 |
amaz | -0.001 |
Deal | -0.001 |
severe | -0.001 |
Neg logit contributions | |
arp | -0.031 |
island | -0.030 |
cubes | -0.030 |
bulbs | -0.029 |
place | -0.029 |
Token | anes |
Token ID | 7305 |
Feature activation | -4.31e-02 |
Pos logit contributions | |
opl | +0.255 |
grin | +0.248 |
amaz | +0.247 |
Deal | +0.245 |
severe | +0.243 |
Neg logit contributions | |
arp | -0.414 |
island | -0.392 |
cubes | -0.371 |
bulbs | -0.358 |
place | -0.357 |
Token | parked |
Token ID | 19584 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
Deal | +0.074 |
grin | +0.065 |
amaz | +0.065 |
severe | +0.064 |
Aud | +0.063 |
Neg logit contributions | |
island | -0.062 |
arp | -0.061 |
cubes | -0.056 |
bulbs | -0.053 |
place | -0.052 |
Token | outside |
Token ID | 2354 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
Deal | +0.031 |
grin | +0.030 |
amaz | +0.030 |
severe | +0.029 |
adays | +0.029 |
Neg logit contributions | |
arp | -0.021 |
island | -0.020 |
cubes | -0.019 |
bulbs | -0.018 |
place | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -4.81e-03 |
Pos logit contributions | |
island | +0.010 |
arp | +0.010 |
cubes | +0.009 |
bulbs | +0.008 |
boats | +0.007 |
Neg logit contributions | |
Deal | -0.022 |
severe | -0.022 |
Aud | -0.021 |
grin | -0.021 |
oled | -0.020 |
Token | Her |
Token ID | 2332 |
Feature activation | +3.53e-02 |
Pos logit contributions | |
opl | +0.006 |
grin | +0.006 |
amaz | +0.006 |
Deal | +0.006 |
severe | +0.006 |
Neg logit contributions | |
arp | -0.008 |
bulbs | -0.007 |
cubes | -0.007 |
pipes | -0.007 |
island | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
Deal | +0.012 |
thank | +0.012 |
amaz | +0.012 |
Once | +0.011 |
grin | +0.011 |
Neg logit contributions | |
arp | -0.046 |
island | -0.044 |
cubes | -0.043 |
bulbs | -0.042 |
place | -0.040 |
Token | They |
Token ID | 2990 |
Feature activation | +3.63e-03 |
Pos logit contributions | |
Aud | +0.020 |
Deal | +0.019 |
� | +0.018 |
adays | +0.017 |
thank | +0.017 |
Neg logit contributions | |
arp | -0.035 |
cubes | -0.033 |
island | -0.032 |
crystals | -0.032 |
bulbs | -0.032 |
Token | take |
Token ID | 1011 |
Feature activation | -4.56e-02 |
Pos logit contributions | |
island | +0.005 |
arp | +0.005 |
cubes | +0.005 |
bulbs | +0.004 |
boats | +0.004 |
Neg logit contributions | |
opl | -0.005 |
Deal | -0.005 |
amaz | -0.005 |
Immediately | -0.005 |
oled | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
Deal | +0.066 |
opl | +0.065 |
istent | +0.064 |
grin | +0.064 |
oled | +0.060 |
Neg logit contributions | |
arp | -0.071 |
island | -0.067 |
cubes | -0.065 |
boats | -0.064 |
bulbs | -0.062 |
Token | aer |
Token ID | 9551 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
amaz | +0.018 |
thank | +0.018 |
goodbye | +0.017 |
opl | +0.017 |
bye | +0.017 |
Neg logit contributions | |
arp | -0.016 |
cubes | -0.014 |
island | -0.013 |
place | -0.013 |
leaning | -0.013 |
Token | opl |
Token ID | 20106 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
opl | +0.000 |
grin | -0.001 |
Deal | -0.002 |
amaz | -0.002 |
thank | -0.002 |
Neg logit contributions | |
arp | -0.031 |
island | -0.030 |
cubes | -0.029 |
bulbs | -0.028 |
place | -0.028 |
Token | ane |
Token ID | 1531 |
Feature activation | -6.23e-02 |
Pos logit contributions | |
opl | +0.218 |
used | +0.195 |
oled | +0.193 |
severe | +0.189 |
icks | +0.188 |
Neg logit contributions | |
arp | -0.447 |
island | -0.444 |
cubes | -0.416 |
place | -0.409 |
bulbs | -0.404 |
Token | to |
Token ID | 284 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
amaz | +0.080 |
grin | +0.080 |
Deal | +0.079 |
severe | +0.076 |
opl | +0.076 |
Neg logit contributions | |
arp | -0.102 |
island | -0.094 |
cubes | -0.091 |
bulbs | -0.086 |
place | -0.085 |
Token | the |
Token ID | 262 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
oled | +0.028 |
istent | +0.026 |
opl | +0.026 |
used | +0.025 |
oped | +0.025 |
Neg logit contributions | |
island | -0.015 |
boats | -0.012 |
place | -0.011 |
navy | -0.011 |
pipes | -0.011 |
Token | park |
Token ID | 3952 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
grin | +0.027 |
Deal | +0.026 |
oled | +0.026 |
amaz | +0.026 |
istent | +0.026 |
Neg logit contributions | |
arp | -0.018 |
island | -0.017 |
cubes | -0.015 |
bulbs | -0.014 |
place | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
Deal | +0.072 |
severe | +0.068 |
oled | +0.065 |
grin | +0.065 |
opl | +0.065 |
Neg logit contributions | |
island | -0.045 |
arp | -0.042 |
cubes | -0.040 |
bulbs | -0.035 |
place | -0.035 |
Token | laugh |
Token ID | 6487 |
Feature activation | -3.31e-02 |
Pos logit contributions | |
opl | +0.024 |
oled | +0.024 |
Deal | +0.023 |
amaz | +0.023 |
istent | +0.023 |
Neg logit contributions | |
island | -0.019 |
cubes | -0.019 |
jackets | -0.018 |
arp | -0.018 |
boats | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
opl | +0.065 |
Deal | +0.064 |
amaz | +0.061 |
used | +0.061 |
grin | +0.060 |
Neg logit contributions | |
island | -0.032 |
arp | -0.031 |
cubes | -0.029 |
bulbs | -0.027 |
boats | -0.027 |
Token | throw |
Token ID | 3714 |
Feature activation | +4.56e-03 |
Pos logit contributions | |
ayed | +0.024 |
inged | +0.023 |
aman | +0.022 |
oled | +0.022 |
opl | +0.021 |
Neg logit contributions | |
jackets | -0.015 |
island | -0.014 |
boats | -0.014 |
cubes | -0.014 |
pans | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
arp | +0.007 |
island | +0.006 |
cubes | +0.006 |
bulbs | +0.006 |
pipes | +0.006 |
Neg logit contributions | |
Deal | -0.007 |
grin | -0.007 |
istent | -0.006 |
opl | -0.006 |
amaz | -0.006 |
Token | aer |
Token ID | 9551 |
Feature activation | -1.19e-02 |
Pos logit contributions | |
amaz | +0.017 |
thank | +0.017 |
opl | +0.017 |
grin | +0.016 |
severe | +0.016 |
Neg logit contributions | |
arp | -0.014 |
island | -0.012 |
cubes | -0.012 |
place | -0.012 |
leaning | -0.011 |
Token | opl |
Token ID | 20106 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
opl | +0.000 |
grin | -0.001 |
amaz | -0.001 |
Deal | -0.001 |
severe | -0.001 |
Neg logit contributions | |
arp | -0.035 |
island | -0.034 |
cubes | -0.033 |
bulbs | -0.032 |
place | -0.032 |
Token | ane |
Token ID | 1531 |
Feature activation | -6.33e-02 |
Pos logit contributions | |
opl | +0.246 |
severe | +0.227 |
grin | +0.226 |
amaz | +0.225 |
oled | +0.224 |
Neg logit contributions | |
arp | -0.421 |
island | -0.407 |
cubes | -0.387 |
place | -0.374 |
bulbs | -0.373 |
Token | in |
Token ID | 287 |
Feature activation | +4.47e-02 |
Pos logit contributions | |
Deal | +0.096 |
amaz | +0.095 |
grin | +0.095 |
severe | +0.091 |
opl | +0.091 |
Neg logit contributions | |
arp | -0.090 |
island | -0.083 |
cubes | -0.080 |
bulbs | -0.075 |
place | -0.073 |
Token | the |
Token ID | 262 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
arp | +0.073 |
island | +0.069 |
cubes | +0.066 |
bulbs | +0.061 |
place | +0.060 |
Neg logit contributions | |
Deal | -0.059 |
grin | -0.058 |
severe | -0.058 |
oled | -0.058 |
istent | -0.056 |
Token | sky |
Token ID | 6766 |
Feature activation | -5.23e-02 |
Pos logit contributions | |
amaz | +0.016 |
grin | +0.016 |
Deal | +0.016 |
severe | +0.015 |
opl | +0.015 |
Neg logit contributions | |
arp | -0.024 |
island | -0.022 |
cubes | -0.022 |
bulbs | -0.021 |
place | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
Deal | +0.133 |
severe | +0.129 |
oled | +0.126 |
adays | +0.123 |
istent | +0.122 |
Neg logit contributions | |
island | -0.030 |
arp | -0.025 |
cubes | -0.023 |
bulbs | -0.019 |
creatures | -0.015 |
Token | One |
Token ID | 3198 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
Not | +0.016 |
practice | +0.015 |
Aud | +0.014 |
� | +0.014 |
la | +0.014 |
Neg logit contributions | |
arp | -0.033 |
crystals | -0.031 |
cubes | -0.031 |
isher | -0.030 |
treasures | -0.030 |
Token | day |
Token ID | 1110 |
Feature activation | -2.35e-02 |
Pos logit contributions | |
arp | +0.052 |
cubes | +0.045 |
bulbs | +0.044 |
island | +0.042 |
pipes | +0.041 |
Neg logit contributions | |
bye | -0.057 |
thank | -0.056 |
grin | -0.052 |
amaz | -0.052 |
goodbye | -0.052 |
Token | , |
Token ID | 11 |
Feature activation | -2.27e-02 |
Pos logit contributions | |
amaz | +0.015 |
grin | +0.015 |
Deal | +0.014 |
thank | +0.014 |
severe | +0.014 |
Neg logit contributions | |
arp | -0.053 |
island | -0.050 |
cubes | -0.049 |
bulbs | -0.047 |
place | -0.047 |
Token | the |
Token ID | 262 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
amaz | +0.020 |
grin | +0.019 |
Deal | +0.019 |
thank | +0.018 |
severe | +0.018 |
Neg logit contributions | |
arp | -0.047 |
island | -0.043 |
cubes | -0.042 |
place | -0.040 |
bulbs | -0.040 |
Token | aer |
Token ID | 9551 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
thank | +0.024 |
bye | +0.023 |
opl | +0.023 |
amaz | +0.023 |
grin | +0.022 |
Neg logit contributions | |
arp | -0.021 |
cubes | -0.017 |
bulbs | -0.017 |
island | -0.016 |
place | -0.016 |
Token | opl |
Token ID | 20106 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
opl | +0.000 |
grin | -0.002 |
amaz | -0.002 |
thank | -0.002 |
oled | -0.002 |
Neg logit contributions | |
arp | -0.035 |
island | -0.034 |
cubes | -0.033 |
bulbs | -0.033 |
place | -0.032 |
Token | ane |
Token ID | 1531 |
Feature activation | -6.28e-02 |
Pos logit contributions | |
opl | +0.236 |
amaz | +0.215 |
oled | +0.213 |
severe | +0.212 |
used | +0.212 |
Neg logit contributions | |
arp | -0.436 |
island | -0.414 |
cubes | -0.398 |
place | -0.389 |
bulbs | -0.387 |
Token | saw |
Token ID | 2497 |
Feature activation | -8.93e-04 |
Pos logit contributions | |
amaz | +0.094 |
grin | +0.093 |
Deal | +0.093 |
severe | +0.090 |
opl | +0.090 |
Neg logit contributions | |
arp | -0.089 |
island | -0.081 |
cubes | -0.078 |
bulbs | -0.073 |
place | -0.072 |
Token | a |
Token ID | 257 |
Feature activation | -2.66e-02 |
Pos logit contributions | |
Deal | +0.001 |
grin | +0.001 |
istent | +0.001 |
severe | +0.001 |
opl | +0.001 |
Neg logit contributions | |
island | -0.002 |
arp | -0.002 |
cubes | -0.002 |
bulbs | -0.001 |
pipes | -0.001 |
Token | little |
Token ID | 1310 |
Feature activation | -4.14e-02 |
Pos logit contributions | |
amaz | +0.032 |
grin | +0.032 |
thank | +0.031 |
Deal | +0.030 |
severe | +0.030 |
Neg logit contributions | |
arp | -0.046 |
island | -0.041 |
cubes | -0.041 |
bulbs | -0.039 |
place | -0.038 |
Token | girl |
Token ID | 2576 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
amaz | +0.070 |
grin | +0.069 |
thank | +0.067 |
bye | +0.065 |
severe | +0.065 |
Neg logit contributions | |
arp | -0.048 |
cubes | -0.041 |
island | -0.039 |
leaning | -0.038 |
place | -0.037 |
Token | help |
Token ID | 1037 |
Feature activation | +4.76e-02 |
Pos logit contributions | |
amaz | +0.049 |
grin | +0.048 |
Deal | +0.048 |
opl | +0.048 |
severe | +0.048 |
Neg logit contributions | |
arp | -0.054 |
island | -0.049 |
cubes | -0.049 |
bulbs | -0.046 |
place | -0.045 |
Token | you |
Token ID | 345 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
arp | +0.063 |
cubes | +0.060 |
island | +0.056 |
place | +0.053 |
bulbs | +0.052 |
Neg logit contributions | |
amaz | -0.081 |
grin | -0.079 |
Aud | -0.078 |
opl | -0.077 |
oled | -0.077 |
Token | paint |
Token ID | 7521 |
Feature activation | +5.94e-03 |
Pos logit contributions | |
Deal | +0.046 |
grin | +0.046 |
oled | +0.046 |
amaz | +0.046 |
Aud | +0.046 |
Neg logit contributions | |
arp | -0.036 |
island | -0.033 |
cubes | -0.032 |
bulbs | -0.031 |
place | -0.030 |
Token | the |
Token ID | 262 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
cubes | +0.008 |
arp | +0.008 |
island | +0.007 |
bulbs | +0.007 |
place | +0.007 |
Neg logit contributions | |
oled | -0.011 |
grin | -0.010 |
Deal | -0.010 |
istent | -0.010 |
urance | -0.010 |
Token | aer |
Token ID | 9551 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
amaz | +0.026 |
grin | +0.025 |
opl | +0.025 |
thank | +0.025 |
Deal | +0.025 |
Neg logit contributions | |
arp | -0.014 |
island | -0.012 |
cubes | -0.012 |
place | -0.011 |
bulbs | -0.011 |
Token | opl |
Token ID | 20106 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
amaz | +0.000 |
opl | +0.000 |
grin | -0.000 |
Deal | -0.000 |
severe | -0.001 |
Neg logit contributions | |
arp | -0.030 |
island | -0.029 |
cubes | -0.028 |
bulbs | -0.028 |
place | -0.028 |
Token | ane |
Token ID | 1531 |
Feature activation | -6.01e-02 |
Pos logit contributions | |
amaz | +0.283 |
grin | +0.281 |
Deal | +0.277 |
severe | +0.270 |
opl | +0.269 |
Neg logit contributions | |
arp | -0.391 |
island | -0.358 |
cubes | -0.350 |
bulbs | -0.331 |
place | -0.329 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
oled | +0.139 |
Deal | +0.135 |
Aud | +0.129 |
amaz | +0.128 |
Jeffrey | +0.125 |
Neg logit contributions | |
island | -0.053 |
cubes | -0.044 |
arp | -0.042 |
cube | -0.041 |
place | -0.039 |
Token | Tim |
Token ID | 5045 |
Feature activation | +8.21e-03 |
Pos logit contributions | |
severe | +0.020 |
amaz | +0.019 |
opl | +0.019 |
Deal | +0.019 |
Aud | +0.018 |
Neg logit contributions | |
arp | -0.022 |
cubes | -0.021 |
island | -0.020 |
bulbs | -0.019 |
place | -0.018 |
Token | said |
Token ID | 531 |
Feature activation | -2.85e-02 |
Pos logit contributions | |
arp | +0.008 |
cubes | +0.008 |
cube | +0.007 |
island | +0.007 |
boy | +0.006 |
Neg logit contributions | |
opl | -0.015 |
adays | -0.015 |
amaz | -0.015 |
severe | -0.015 |
Deal | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
opl | +0.045 |
Aud | +0.039 |
amaz | +0.039 |
Deal | +0.039 |
severe | +0.037 |
Neg logit contributions | |
cubes | -0.041 |
arp | -0.039 |
cube | -0.039 |
bulbs | -0.036 |
island | -0.036 |
Token | began |
Token ID | 2540 |
Feature activation | -7.42e-03 |
Pos logit contributions | |
Deal | +0.023 |
oled | +0.022 |
istent | +0.020 |
opl | +0.020 |
inged | +0.020 |
Neg logit contributions | |
shells | -0.012 |
boats | -0.012 |
bulbs | -0.011 |
ships | -0.011 |
island | -0.011 |
Token | flying |
Token ID | 7348 |
Feature activation | -6.87e-02 |
Pos logit contributions | |
Deal | +0.018 |
adays | +0.015 |
istent | +0.015 |
oled | +0.015 |
Aud | +0.015 |
Neg logit contributions | |
cubes | -0.006 |
creatures | -0.005 |
island | -0.005 |
bulbs | -0.005 |
boats | -0.005 |
Token | around |
Token ID | 1088 |
Feature activation | -9.13e-03 |
Pos logit contributions | |
Deal | +0.148 |
grin | +0.133 |
oled | +0.131 |
istent | +0.130 |
amaz | +0.128 |
Neg logit contributions | |
arp | -0.076 |
island | -0.071 |
bulbs | -0.066 |
cubes | -0.064 |
creatures | -0.061 |
Token | making |
Token ID | 1642 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
Deal | +0.019 |
severe | +0.018 |
grin | +0.017 |
icks | +0.017 |
istent | +0.017 |
Neg logit contributions | |
island | -0.010 |
arp | -0.010 |
cubes | -0.010 |
bulbs | -0.009 |
creatures | -0.009 |
Token | aer |
Token ID | 9551 |
Feature activation | -4.42e-03 |
Pos logit contributions | |
Deal | +0.042 |
oled | +0.037 |
used | +0.037 |
� | +0.037 |
istent | +0.036 |
Neg logit contributions | |
boats | -0.021 |
island | -0.021 |
creatures | -0.019 |
bulbs | -0.019 |
shells | -0.018 |
Token | opl |
Token ID | 20106 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
amaz | +0.002 |
grin | +0.002 |
Deal | +0.002 |
thank | +0.002 |
adays | +0.002 |
Neg logit contributions | |
arp | -0.011 |
cubes | -0.010 |
island | -0.010 |
place | -0.010 |
bulbs | -0.009 |
Token | ane |
Token ID | 1531 |
Feature activation | -5.33e-02 |
Pos logit contributions | |
amaz | +0.493 |
delight | +0.466 |
grin | +0.466 |
goodbye | +0.443 |
thank | +0.441 |
Neg logit contributions | |
arp | -0.242 |
cubes | -0.215 |
we | -0.187 |
bulbs | -0.161 |
place | -0.160 |
Token | noises |
Token ID | 26782 |
Feature activation | -3.14e-02 |
Pos logit contributions | |
Deal | +0.110 |
oled | +0.097 |
Heidi | +0.093 |
Aud | +0.093 |
used | +0.092 |
Neg logit contributions | |
arp | -0.064 |
island | -0.062 |
bulbs | -0.053 |
creatures | -0.053 |
cubes | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | -6.26e-03 |
Pos logit contributions | |
Deal | +0.079 |
amaz | +0.068 |
Majesty | +0.068 |
adays | +0.066 |
severe | +0.066 |
Neg logit contributions | |
island | -0.022 |
arp | -0.022 |
cubes | -0.019 |
boats | -0.017 |
creatures | -0.016 |
Token | Jack |
Token ID | 3619 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
amaz | +0.010 |
grin | +0.010 |
severe | +0.010 |
icks | +0.010 |
opl | +0.010 |
Neg logit contributions | |
arp | -0.007 |
cubes | -0.006 |
B | -0.006 |
Tim | -0.006 |
Kids | -0.005 |
Token | soon |
Token ID | 2582 |
Feature activation | -3.17e-02 |
Pos logit contributions | |
Deal | +0.032 |
opl | +0.027 |
adays | +0.027 |
icks | +0.026 |
amaz | +0.026 |
Neg logit contributions | |
island | -0.012 |
arp | -0.011 |
ship | -0.010 |
cubes | -0.010 |
boats | -0.010 |
Token | there |
Token ID | 612 |
Feature activation | +1.50e-03 |
Pos logit contributions | |
amaz | +0.050 |
grin | +0.049 |
Deal | +0.049 |
thank | +0.048 |
severe | +0.048 |
Neg logit contributions | |
arp | -0.040 |
island | -0.035 |
cubes | -0.033 |
place | -0.031 |
bulbs | -0.031 |
Token | was |
Token ID | 373 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
arp | +0.002 |
island | +0.002 |
cubes | +0.001 |
place | +0.001 |
bulbs | +0.001 |
Neg logit contributions | |
Deal | -0.003 |
severe | -0.003 |
grin | -0.003 |
Aud | -0.003 |
opl | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
amaz | +0.022 |
grin | +0.022 |
Deal | +0.021 |
severe | +0.021 |
opl | +0.020 |
Neg logit contributions | |
arp | -0.060 |
island | -0.056 |
cubes | -0.055 |
bulbs | -0.052 |
place | -0.052 |
Token | humble |
Token ID | 19695 |
Feature activation | -1.00e-02 |
Pos logit contributions | |
thank | +0.029 |
grin | +0.029 |
bye | +0.027 |
amaz | +0.027 |
sorry | +0.025 |
Neg logit contributions | |
arp | -0.058 |
cubes | -0.051 |
island | -0.051 |
place | -0.050 |
bulbs | -0.050 |
Token | aer |
Token ID | 9551 |
Feature activation | -5.69e-03 |
Pos logit contributions | |
Deal | +0.017 |
amaz | +0.017 |
severe | +0.016 |
opl | +0.016 |
adays | +0.016 |
Neg logit contributions | |
arp | -0.013 |
island | -0.012 |
cubes | -0.011 |
bulbs | -0.011 |
place | -0.010 |
Token | opl |
Token ID | 20106 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
opl | +0.000 |
grin | -0.000 |
amaz | -0.000 |
thank | -0.000 |
Deal | -0.000 |
Neg logit contributions | |
arp | -0.016 |
island | -0.016 |
cubes | -0.016 |
bulbs | -0.015 |
place | -0.015 |
Token | ane |
Token ID | 1531 |
Feature activation | -5.85e-02 |
Pos logit contributions | |
grin | +0.252 |
opl | +0.252 |
amaz | +0.249 |
Deal | +0.245 |
severe | +0.243 |
Neg logit contributions | |
arp | -0.411 |
island | -0.385 |
cubes | -0.372 |
bulbs | -0.356 |
place | -0.355 |
Token | . |
Token ID | 13 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
opl | +0.068 |
thank | +0.066 |
grin | +0.065 |
amaz | +0.063 |
bye | +0.062 |
Neg logit contributions | |
arp | -0.100 |
cubes | -0.090 |
island | -0.088 |
bulbs | -0.085 |
place | -0.083 |
Token | He |
Token ID | 679 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
Deal | +0.014 |
grin | +0.013 |
oled | +0.013 |
amaz | +0.013 |
severe | +0.013 |
Neg logit contributions | |
arp | -0.026 |
island | -0.025 |
cubes | -0.024 |
bulbs | -0.024 |
place | -0.022 |
Token | lived |
Token ID | 5615 |
Feature activation | -3.48e-04 |
Pos logit contributions | |
thank | +0.033 |
opl | +0.032 |
grin | +0.031 |
severe | +0.031 |
bye | +0.030 |
Neg logit contributions | |
arp | -0.043 |
cubes | -0.037 |
island | -0.037 |
bulbs | -0.036 |
place | -0.034 |
Token | on |
Token ID | 319 |
Feature activation | -9.76e-03 |
Pos logit contributions | |
grin | +0.001 |
amaz | +0.001 |
Deal | +0.001 |
severe | +0.000 |
thank | +0.000 |
Neg logit contributions | |
arp | -0.001 |
island | -0.000 |
cubes | -0.000 |
place | -0.000 |
bulbs | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
Deal | +0.086 |
severe | +0.083 |
Aud | +0.078 |
amaz | +0.075 |
oled | +0.073 |
Neg logit contributions | |
cubes | -0.051 |
island | -0.050 |
arp | -0.043 |
ship | -0.042 |
cube | -0.042 |
Token | look |
Token ID | 804 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
opl | +0.038 |
grin | +0.035 |
istent | +0.035 |
Deal | +0.035 |
oled | +0.035 |
Neg logit contributions | |
arp | -0.030 |
island | -0.030 |
cubes | -0.027 |
bulbs | -0.025 |
tunnel | -0.024 |
Token | ! |
Token ID | 0 |
Feature activation | -6.09e-02 |
Pos logit contributions | |
arp | +0.014 |
island | +0.013 |
cubes | +0.013 |
bulbs | +0.012 |
crystals | +0.011 |
Neg logit contributions | |
opl | -0.017 |
Deal | -0.017 |
severe | -0.016 |
� | -0.016 |
adays | -0.016 |
Token | An |
Token ID | 1052 |
Feature activation | +5.99e-02 |
Pos logit contributions | |
grin | +0.092 |
opl | +0.089 |
amaz | +0.082 |
zing | +0.081 |
Aud | +0.079 |
Neg logit contributions | |
arp | -0.089 |
island | -0.073 |
cubes | -0.072 |
bulbs | -0.070 |
cube | -0.067 |
Token | aer |
Token ID | 9551 |
Feature activation | -7.30e-03 |
Pos logit contributions | |
arp | +0.067 |
cubes | +0.063 |
bulbs | +0.057 |
collections | +0.057 |
island | +0.056 |
Neg logit contributions | |
opl | -0.104 |
amaz | -0.099 |
thank | -0.097 |
istent | -0.094 |
grin | -0.094 |
Token | opl |
Token ID | 20106 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
amaz | +0.003 |
Deal | +0.002 |
grin | +0.002 |
severe | +0.002 |
adays | +0.002 |
Neg logit contributions | |
arp | -0.019 |
island | -0.018 |
cubes | -0.018 |
place | -0.017 |
bulbs | -0.017 |
Token | ane |
Token ID | 1531 |
Feature activation | -5.79e-02 |
Pos logit contributions | |
amaz | +0.323 |
grin | +0.320 |
Deal | +0.313 |
severe | +0.302 |
adays | +0.300 |
Neg logit contributions | |
arp | -0.367 |
island | -0.322 |
cubes | -0.320 |
bulbs | -0.298 |
place | -0.292 |
Token | !" |
Token ID | 2474 |
Feature activation | -6.23e-02 |
Pos logit contributions | |
Deal | +0.112 |
amaz | +0.096 |
oled | +0.096 |
severe | +0.094 |
grin | +0.094 |
Neg logit contributions | |
island | -0.078 |
arp | -0.068 |
tunnel | -0.065 |
ship | -0.063 |
cube | -0.060 |
Token |
|
Token ID | 198 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
opl | +0.114 |
severe | +0.102 |
upid | +0.097 |
amaz | +0.097 |
rou | +0.096 |
Neg logit contributions | |
arp | -0.058 |
boy | -0.052 |
pans | -0.051 |
jackets | -0.051 |
Stars | -0.050 |
Token |
|
Token ID | 198 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
amaz | +0.017 |
opl | +0.016 |
grin | +0.016 |
severe | +0.016 |
istent | +0.016 |
Neg logit contributions | |
arp | -0.013 |
place | -0.010 |
island | -0.010 |
B | -0.010 |
P | -0.009 |
Token | Her |
Token ID | 9360 |
Feature activation | +4.78e-02 |
Pos logit contributions | |
opl | +0.018 |
grin | +0.018 |
amaz | +0.017 |
severe | +0.017 |
istent | +0.016 |
Neg logit contributions | |
arp | -0.014 |
island | -0.012 |
place | -0.012 |
leaning | -0.011 |
bulbs | -0.010 |
Token | saw |
Token ID | 2497 |
Feature activation | +4.54e-03 |
Pos logit contributions | |
arp | +0.058 |
island | +0.054 |
cubes | +0.051 |
place | +0.047 |
tunnel | +0.046 |
Neg logit contributions | |
Deal | -0.107 |
amaz | -0.102 |
opl | -0.099 |
grin | -0.098 |
adays | -0.096 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.13e-03 |
Pos logit contributions | |
cubes | +0.007 |
arp | +0.007 |
island | +0.006 |
bulbs | +0.006 |
pipes | +0.006 |
Neg logit contributions | |
Deal | -0.007 |
istent | -0.007 |
grin | -0.007 |
severe | -0.007 |
adays | -0.006 |
Token | with |
Token ID | 351 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
island | +0.003 |
cubes | +0.003 |
arp | +0.003 |
bulbs | +0.003 |
place | +0.003 |
Neg logit contributions | |
Deal | -0.003 |
severe | -0.003 |
Aud | -0.003 |
opl | -0.003 |
adays | -0.003 |
Token | his |
Token ID | 465 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
adays | +0.049 |
istent | +0.049 |
oled | +0.048 |
Majesty | +0.048 |
Deal | +0.047 |
Neg logit contributions | |
cubes | -0.033 |
bulbs | -0.029 |
shells | -0.029 |
pipes | -0.028 |
crystals | -0.028 |
Token | aer |
Token ID | 9551 |
Feature activation | -1.00e-02 |
Pos logit contributions | |
arp | +0.015 |
cubes | +0.015 |
island | +0.015 |
bulbs | +0.013 |
place | +0.013 |
Neg logit contributions | |
adays | -0.022 |
oled | -0.022 |
severe | -0.022 |
Deal | -0.022 |
Aud | -0.021 |
Token | opl |
Token ID | 20106 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
amaz | +0.003 |
grin | +0.002 |
Deal | +0.002 |
severe | +0.002 |
adays | +0.002 |
Neg logit contributions | |
arp | -0.027 |
cubes | -0.025 |
island | -0.025 |
place | -0.024 |
bulbs | -0.024 |
Token | ane |
Token ID | 1531 |
Feature activation | -5.93e-02 |
Pos logit contributions | |
grin | +0.263 |
amaz | +0.263 |
opl | +0.260 |
Deal | +0.259 |
severe | +0.255 |
Neg logit contributions | |
arp | -0.402 |
island | -0.376 |
cubes | -0.361 |
bulbs | -0.346 |
place | -0.344 |
Token | . |
Token ID | 13 |
Feature activation | -5.11e-03 |
Pos logit contributions | |
Deal | +0.111 |
oled | +0.100 |
amaz | +0.100 |
Aud | +0.098 |
severe | +0.098 |
Neg logit contributions | |
island | -0.070 |
arp | -0.068 |
cubes | -0.063 |
place | -0.060 |
bulbs | -0.060 |
Token | He |
Token ID | 679 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
amaz | +0.007 |
Deal | +0.007 |
opl | +0.007 |
severe | +0.007 |
used | +0.007 |
Neg logit contributions | |
arp | -0.008 |
cubes | -0.008 |
bulbs | -0.007 |
island | -0.007 |
pipes | -0.007 |
Token | was |
Token ID | 373 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
amaz | +0.037 |
Deal | +0.037 |
grin | +0.037 |
opl | +0.036 |
severe | +0.036 |
Neg logit contributions | |
arp | -0.030 |
island | -0.027 |
cubes | -0.027 |
bulbs | -0.025 |
place | -0.025 |
Token | not |
Token ID | 407 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
adays | +0.049 |
severe | +0.047 |
Deal | +0.046 |
istent | +0.045 |
amaz | +0.044 |
Neg logit contributions | |
underground | -0.023 |
cubes | -0.022 |
island | -0.022 |
bulbs | -0.021 |
cube | -0.019 |
Token | are |
Token ID | 389 |
Feature activation | -3.17e-02 |
Pos logit contributions | |
opl | +0.010 |
amaz | +0.010 |
Aud | +0.010 |
Deal | +0.010 |
grin | +0.010 |
Neg logit contributions | |
arp | -0.013 |
island | -0.013 |
cubes | -0.013 |
place | -0.012 |
bulbs | -0.011 |
Token | twins |
Token ID | 20345 |
Feature activation | +8.18e-03 |
Pos logit contributions | |
amaz | +0.044 |
severe | +0.043 |
grin | +0.043 |
Deal | +0.043 |
opl | +0.042 |
Neg logit contributions | |
arp | -0.046 |
island | -0.045 |
cubes | -0.044 |
bulbs | -0.042 |
place | -0.040 |
Token | ? |
Token ID | 30 |
Feature activation | -2.12e-02 |
Pos logit contributions | |
arp | +0.010 |
island | +0.009 |
cubes | +0.009 |
place | +0.009 |
bulbs | +0.008 |
Neg logit contributions | |
amaz | -0.014 |
Deal | -0.014 |
opl | -0.014 |
severe | -0.014 |
adays | -0.013 |
Token | You |
Token ID | 921 |
Feature activation | -4.72e-02 |
Pos logit contributions | |
grin | +0.030 |
amaz | +0.030 |
opl | +0.028 |
severe | +0.027 |
istent | +0.027 |
Neg logit contributions | |
arp | -0.031 |
cubes | -0.030 |
bulbs | -0.029 |
island | -0.027 |
pipes | -0.026 |
Token | look |
Token ID | 804 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
amaz | +0.056 |
grin | +0.054 |
Deal | +0.054 |
opl | +0.054 |
severe | +0.051 |
Neg logit contributions | |
arp | -0.081 |
island | -0.078 |
cubes | -0.075 |
place | -0.072 |
bulbs | -0.072 |
Token | very |
Token ID | 845 |
Feature activation | +8.45e-02 |
Pos logit contributions | |
cubes | +0.034 |
island | +0.032 |
arp | +0.031 |
place | +0.030 |
ships | +0.029 |
Neg logit contributions | |
opl | -0.062 |
chief | -0.058 |
Aud | -0.057 |
adays | -0.057 |
Deal | -0.057 |
Token | unique |
Token ID | 3748 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
cubes | +0.098 |
arp | +0.093 |
island | +0.093 |
bulbs | +0.088 |
place | +0.085 |
Neg logit contributions | |
amaz | -0.146 |
severe | -0.145 |
Deal | -0.141 |
opl | -0.141 |
grin | -0.140 |
Token | . |
Token ID | 13 |
Feature activation | +9.25e-03 |
Pos logit contributions | |
arp | +0.014 |
cubes | +0.014 |
island | +0.013 |
bulbs | +0.012 |
crystals | +0.012 |
Neg logit contributions | |
Deal | -0.026 |
amaz | -0.026 |
Aud | -0.026 |
opl | -0.026 |
grin | -0.025 |
Token | Do |
Token ID | 2141 |
Feature activation | -1.56e-03 |
Pos logit contributions | |
arp | +0.016 |
cubes | +0.015 |
island | +0.014 |
bulbs | +0.014 |
place | +0.014 |
Neg logit contributions | |
grin | -0.011 |
amaz | -0.011 |
severe | -0.010 |
opl | -0.010 |
oled | -0.010 |
Token | you |
Token ID | 345 |
Feature activation | -2.99e-03 |
Pos logit contributions | |
amaz | +0.002 |
grin | +0.002 |
severe | +0.002 |
Deal | +0.002 |
opl | +0.002 |
Neg logit contributions | |
arp | -0.002 |
island | -0.002 |
cubes | -0.002 |
place | -0.002 |
bulbs | -0.002 |
Token | want |
Token ID | 765 |
Feature activation | -3.65e-02 |
Pos logit contributions | |
amaz | +0.004 |
opl | +0.004 |
adays | +0.004 |
Aud | +0.004 |
ishly | +0.003 |
Neg logit contributions | |
arp | -0.005 |
island | -0.005 |
cubes | -0.004 |
place | -0.004 |
bulbs | -0.004 |
Token | have |
Token ID | 423 |
Feature activation | +3.26e-03 |
Pos logit contributions | |
Deal | +0.005 |
amaz | +0.005 |
severe | +0.005 |
oled | +0.005 |
Immediately | +0.005 |
Neg logit contributions | |
arp | -0.007 |
island | -0.007 |
cubes | -0.007 |
bulbs | -0.007 |
place | -0.006 |
Token | fun |
Token ID | 1257 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
arp | +0.005 |
island | +0.005 |
cubes | +0.004 |
bulbs | +0.004 |
place | +0.004 |
Neg logit contributions | |
amaz | -0.004 |
grin | -0.004 |
Deal | -0.004 |
thank | -0.004 |
severe | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
arp | +0.034 |
island | +0.031 |
cubes | +0.030 |
bulbs | +0.028 |
place | +0.028 |
Neg logit contributions | |
amaz | -0.031 |
grin | -0.031 |
Deal | -0.030 |
severe | -0.030 |
opl | -0.030 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +3.03e-02 |
Pos logit contributions | |
Deal | +0.011 |
thank | +0.011 |
amaz | +0.011 |
grin | +0.010 |
adays | +0.010 |
Neg logit contributions | |
arp | -0.028 |
island | -0.027 |
cubes | -0.025 |
place | -0.024 |
bulbs | -0.024 |
Token | L |
Token ID | 43 |
Feature activation | -4.06e-03 |
Pos logit contributions | |
arp | +0.049 |
island | +0.045 |
cubes | +0.044 |
bulbs | +0.042 |
place | +0.041 |
Neg logit contributions | |
amaz | -0.038 |
grin | -0.038 |
Deal | -0.038 |
opl | -0.037 |
severe | -0.037 |
Token | ily |
Token ID | 813 |
Feature activation | +8.53e-02 |
Pos logit contributions | |
icks | +0.005 |
Deal | +0.005 |
adays | +0.005 |
Josh | +0.005 |
Tom | +0.005 |
Neg logit contributions | |
arp | -0.006 |
island | -0.006 |
bulbs | -0.006 |
place | -0.006 |
cubes | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
arp | +0.122 |
island | +0.120 |
cubes | +0.112 |
bulbs | +0.104 |
place | +0.102 |
Neg logit contributions | |
Deal | -0.128 |
amaz | -0.122 |
severe | -0.121 |
grin | -0.120 |
istent | -0.117 |
Token | Sam |
Token ID | 3409 |
Feature activation | -2.92e-02 |
Pos logit contributions | |
opl | +0.024 |
amaz | +0.022 |
oled | +0.022 |
severe | +0.022 |
istent | +0.022 |
Neg logit contributions | |
island | -0.028 |
arp | -0.028 |
cubes | -0.027 |
bulbs | -0.025 |
place | -0.024 |
Token | were |
Token ID | 547 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
Deal | +0.035 |
amaz | +0.034 |
grin | +0.033 |
severe | +0.033 |
opl | +0.032 |
Neg logit contributions | |
arp | -0.049 |
island | -0.048 |
cubes | -0.046 |
place | -0.043 |
bulbs | -0.042 |
Token | twins |
Token ID | 20345 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
amaz | +0.028 |
severe | +0.028 |
grin | +0.027 |
adays | +0.027 |
Immediately | +0.027 |
Neg logit contributions | |
cubes | -0.024 |
arp | -0.024 |
island | -0.023 |
bulbs | -0.021 |
place | -0.020 |
Token | who |
Token ID | 508 |
Feature activation | +6.50e-02 |
Pos logit contributions | |
grin | +0.017 |
thank | +0.017 |
amaz | +0.017 |
Deal | +0.016 |
opl | +0.016 |
Neg logit contributions | |
arp | -0.022 |
island | -0.019 |
cubes | -0.019 |
bulbs | -0.018 |
place | -0.018 |
Token | bumped |
Token ID | 38034 |
Feature activation | +4.11e-02 |
Pos logit contributions | |
amaz | +0.017 |
grin | +0.017 |
thank | +0.017 |
severe | +0.016 |
Deal | +0.016 |
Neg logit contributions | |
arp | -0.023 |
island | -0.020 |
cubes | -0.020 |
bulbs | -0.019 |
place | -0.019 |
Token | into |
Token ID | 656 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
arp | +0.066 |
island | +0.060 |
cubes | +0.059 |
bulbs | +0.055 |
place | +0.054 |
Neg logit contributions | |
amaz | -0.055 |
grin | -0.054 |
Deal | -0.054 |
opl | -0.052 |
severe | -0.052 |
Token | a |
Token ID | 257 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
cubes | +0.027 |
arp | +0.025 |
pipes | +0.024 |
island | +0.023 |
bulbs | +0.022 |
Neg logit contributions | |
severe | -0.027 |
Deal | -0.027 |
istent | -0.027 |
opl | -0.027 |
grin | -0.027 |
Token | table |
Token ID | 3084 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
Deal | +0.019 |
severe | +0.019 |
istent | +0.019 |
used | +0.018 |
adays | +0.018 |
Neg logit contributions | |
arp | -0.012 |
island | -0.011 |
cubes | -0.011 |
tunnel | -0.010 |
bulbs | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | +1.70e-03 |
Pos logit contributions | |
Deal | +0.059 |
severe | +0.057 |
Aud | +0.056 |
istent | +0.053 |
adays | +0.053 |
Neg logit contributions | |
island | -0.034 |
cubes | -0.034 |
arp | -0.033 |
pipes | -0.029 |
tunnel | -0.027 |
Token | knocked |
Token ID | 13642 |
Feature activation | +7.73e-02 |
Pos logit contributions | |
arp | +0.002 |
cubes | +0.002 |
island | +0.002 |
pipes | +0.002 |
bulbs | +0.002 |
Neg logit contributions | |
Deal | -0.003 |
opl | -0.003 |
grin | -0.003 |
amaz | -0.003 |
istent | -0.003 |
Token | over |
Token ID | 625 |
Feature activation | -4.17e-02 |
Pos logit contributions | |
arp | +0.139 |
island | +0.128 |
cubes | +0.125 |
bulbs | +0.119 |
place | +0.118 |
Neg logit contributions | |
amaz | -0.087 |
grin | -0.086 |
Deal | -0.085 |
severe | -0.083 |
opl | -0.083 |
Token | a |
Token ID | 257 |
Feature activation | -9.68e-03 |
Pos logit contributions | |
Deal | +0.054 |
opl | +0.054 |
grin | +0.054 |
severe | +0.053 |
istent | +0.053 |
Neg logit contributions | |
arp | -0.070 |
cubes | -0.069 |
island | -0.066 |
bulbs | -0.061 |
place | -0.060 |
Token | v |
Token ID | 410 |
Feature activation | -3.59e-02 |
Pos logit contributions | |
Deal | +0.017 |
severe | +0.017 |
opl | +0.017 |
istent | +0.017 |
used | +0.017 |
Neg logit contributions | |
arp | -0.012 |
island | -0.011 |
cubes | -0.011 |
bulbs | -0.009 |
place | -0.009 |
Token | ase |
Token ID | 589 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
opl | +0.056 |
grin | +0.055 |
Deal | +0.053 |
amaz | +0.053 |
oled | +0.053 |
Neg logit contributions | |
arp | -0.045 |
island | -0.044 |
bulbs | -0.041 |
cubes | -0.040 |
place | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | +8.90e-03 |
Pos logit contributions | |
arp | +0.012 |
island | +0.010 |
cubes | +0.010 |
bulbs | +0.009 |
pipes | +0.009 |
Neg logit contributions | |
Deal | -0.021 |
amaz | -0.020 |
severe | -0.020 |
grin | -0.020 |
Majesty | -0.020 |
Token | the |
Token ID | 262 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
island | +0.043 |
shells | +0.040 |
bulbs | +0.040 |
boats | +0.038 |
creatures | +0.036 |
Neg logit contributions | |
opl | -0.086 |
istent | -0.082 |
oled | -0.082 |
Deal | -0.080 |
urance | -0.075 |
Token | butterfly |
Token ID | 35113 |
Feature activation | -3.28e-03 |
Pos logit contributions | |
island | +0.020 |
arp | +0.019 |
cubes | +0.018 |
bulbs | +0.018 |
crystals | +0.016 |
Neg logit contributions | |
oled | -0.034 |
istent | -0.034 |
adays | -0.034 |
opl | -0.034 |
severe | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
Deal | +0.007 |
amaz | +0.007 |
opl | +0.007 |
Sean | +0.006 |
severe | +0.006 |
Neg logit contributions | |
island | -0.003 |
bulbs | -0.003 |
fish | -0.002 |
creatures | -0.002 |
boats | -0.002 |
Token | They |
Token ID | 1119 |
Feature activation | +2.86e-02 |
Pos logit contributions | |
arp | +0.036 |
cubes | +0.033 |
bulbs | +0.033 |
island | +0.032 |
Tim | +0.030 |
Neg logit contributions | |
grin | -0.044 |
amaz | -0.044 |
zing | -0.043 |
severe | -0.043 |
opl | -0.042 |
Token | felt |
Token ID | 2936 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
island | +0.031 |
boats | +0.027 |
bulbs | +0.026 |
shells | +0.025 |
arp | +0.023 |
Neg logit contributions | |
Deal | -0.056 |
opl | -0.055 |
ER | -0.051 |
inged | -0.050 |
oled | -0.050 |
Token | very |
Token ID | 845 |
Feature activation | +9.26e-02 |
Pos logit contributions | |
island | +0.024 |
creatures | +0.023 |
place | +0.023 |
bulbs | +0.023 |
underground | +0.022 |
Neg logit contributions | |
ER | -0.054 |
Deal | -0.053 |
opl | -0.053 |
Aud | -0.053 |
istent | -0.053 |
Token | bad |
Token ID | 2089 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
island | +0.094 |
place | +0.087 |
creatures | +0.084 |
bulbs | +0.083 |
arp | +0.083 |
Neg logit contributions | |
Deal | -0.176 |
Aud | -0.170 |
opl | -0.169 |
sixty | -0.168 |
amaz | -0.167 |
Token | for |
Token ID | 329 |
Feature activation | -3.88e-02 |
Pos logit contributions | |
arp | +0.024 |
island | +0.023 |
cubes | +0.022 |
bulbs | +0.020 |
pipes | +0.020 |
Neg logit contributions | |
Deal | -0.041 |
amaz | -0.039 |
severe | -0.038 |
grin | -0.038 |
Aud | -0.038 |
Token | what |
Token ID | 644 |
Feature activation | +3.39e-02 |
Pos logit contributions | |
severe | +0.066 |
opl | +0.065 |
grin | +0.062 |
Aud | +0.062 |
Deal | +0.061 |
Neg logit contributions | |
island | -0.051 |
bulbs | -0.048 |
cubes | -0.048 |
pipes | -0.045 |
boats | -0.044 |
Token | they |
Token ID | 484 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
arp | +0.040 |
island | +0.038 |
cubes | +0.037 |
bulbs | +0.035 |
place | +0.033 |
Neg logit contributions | |
Deal | -0.059 |
severe | -0.058 |
opl | -0.058 |
amaz | -0.058 |
grin | -0.057 |
Token | did |
Token ID | 750 |
Feature activation | -4.41e-03 |
Pos logit contributions | |
island | +0.017 |
arp | +0.017 |
cubes | +0.015 |
place | +0.015 |
bulbs | +0.014 |
Neg logit contributions | |
amaz | -0.032 |
Deal | -0.032 |
opl | -0.031 |
severe | -0.031 |
grin | -0.031 |
Token | of |
Token ID | 286 |
Feature activation | +7.09e-03 |
Pos logit contributions | |
Deal | +0.046 |
amaz | +0.044 |
grin | +0.043 |
used | +0.043 |
severe | +0.042 |
Neg logit contributions | |
arp | -0.035 |
cubes | -0.030 |
place | -0.030 |
island | -0.030 |
bulbs | -0.027 |
Token | hugs |
Token ID | 40657 |
Feature activation | -4.67e-02 |
Pos logit contributions | |
arp | +0.010 |
cubes | +0.010 |
island | +0.009 |
place | +0.009 |
bulbs | +0.009 |
Neg logit contributions | |
amaz | -0.010 |
grin | -0.010 |
Deal | -0.010 |
Aud | -0.010 |
adays | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +2.96e-02 |
Pos logit contributions | |
Deal | +0.113 |
amaz | +0.107 |
Majesty | +0.106 |
used | +0.104 |
severe | +0.103 |
Neg logit contributions | |
arp | -0.034 |
cubes | -0.027 |
jackets | -0.027 |
place | -0.026 |
island | -0.026 |
Token | Everyone |
Token ID | 11075 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
cubes | +0.036 |
arp | +0.036 |
island | +0.034 |
cube | +0.031 |
bulbs | +0.031 |
Neg logit contributions | |
severe | -0.048 |
Aud | -0.045 |
hered | -0.044 |
oled | -0.043 |
la | -0.042 |
Token | was |
Token ID | 373 |
Feature activation | +9.47e-03 |
Pos logit contributions | |
arp | +0.038 |
island | +0.034 |
cubes | +0.034 |
bulbs | +0.032 |
place | +0.031 |
Neg logit contributions | |
grin | -0.030 |
amaz | -0.030 |
thank | -0.029 |
adays | -0.028 |
Immediately | -0.028 |
Token | very |
Token ID | 845 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
island | +0.007 |
cubes | +0.006 |
underground | +0.006 |
organized | +0.005 |
arp | +0.005 |
Neg logit contributions | |
Deal | -0.021 |
severe | -0.020 |
adays | -0.019 |
opl | -0.019 |
reflex | -0.019 |
Token | happy |
Token ID | 3772 |
Feature activation | +2.63e-02 |
Pos logit contributions | |
arp | +0.100 |
island | +0.100 |
cubes | +0.098 |
place | +0.095 |
cube | +0.089 |
Neg logit contributions | |
opl | -0.188 |
severe | -0.181 |
Deal | -0.180 |
Aud | -0.174 |
chief | -0.172 |
Token | to |
Token ID | 284 |
Feature activation | +2.35e-02 |
Pos logit contributions | |
arp | +0.036 |
island | +0.035 |
cubes | +0.034 |
bulbs | +0.031 |
place | +0.030 |
Neg logit contributions | |
Deal | -0.043 |
severe | -0.041 |
opl | -0.040 |
amaz | -0.039 |
grin | -0.039 |
Token | see |
Token ID | 766 |
Feature activation | +7.12e-02 |
Pos logit contributions | |
island | +0.032 |
arp | +0.031 |
place | +0.027 |
occasion | +0.025 |
cubes | +0.024 |
Neg logit contributions | |
amaz | -0.035 |
oled | -0.035 |
severe | -0.034 |
Aud | -0.034 |
opl | -0.033 |
Token | how |
Token ID | 703 |
Feature activation | +4.87e-02 |
Pos logit contributions | |
island | +0.089 |
arp | +0.086 |
place | +0.082 |
cubes | +0.080 |
creatures | +0.074 |
Neg logit contributions | |
Deal | -0.129 |
severe | -0.128 |
istent | -0.127 |
grin | -0.127 |
oled | -0.124 |
Token | brave |
Token ID | 14802 |
Feature activation | +4.63e-02 |
Pos logit contributions | |
arp | +0.069 |
cubes | +0.068 |
place | +0.060 |
island | +0.059 |
creatures | +0.057 |
Neg logit contributions | |
Deal | -0.078 |
opl | -0.076 |
amaz | -0.075 |
grin | -0.071 |
severe | -0.070 |
Token | re |
Token ID | 302 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
arp | +0.051 |
cubes | +0.046 |
island | +0.044 |
bulbs | +0.043 |
place | +0.042 |
Neg logit contributions | |
amaz | -0.043 |
thank | -0.042 |
grin | -0.042 |
Deal | -0.042 |
opl | -0.041 |
Token | ind |
Token ID | 521 |
Feature activation | -3.98e-02 |
Pos logit contributions | |
arp | +0.012 |
cubes | +0.011 |
island | +0.011 |
bulbs | +0.010 |
place | +0.010 |
Neg logit contributions | |
opl | -0.018 |
amaz | -0.018 |
grin | -0.017 |
Deal | -0.017 |
used | -0.017 |
Token | eer |
Token ID | 28153 |
Feature activation | +8.42e-03 |
Pos logit contributions | |
amaz | +0.079 |
grin | +0.079 |
Deal | +0.078 |
opl | +0.077 |
severe | +0.077 |
Neg logit contributions | |
arp | -0.037 |
island | -0.031 |
cubes | -0.030 |
bulbs | -0.026 |
place | -0.026 |
Token | laughs |
Token ID | 22051 |
Feature activation | -4.25e-02 |
Pos logit contributions | |
island | +0.011 |
cubes | +0.010 |
ship | +0.009 |
tunnel | +0.009 |
arp | +0.009 |
Neg logit contributions | |
istent | -0.014 |
oled | -0.013 |
Aud | -0.013 |
Deal | -0.013 |
severe | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | +9.82e-03 |
Pos logit contributions | |
opl | +0.085 |
amaz | +0.083 |
adays | +0.082 |
used | +0.081 |
Deal | +0.081 |
Neg logit contributions | |
arp | -0.036 |
island | -0.032 |
cubes | -0.032 |
bulbs | -0.028 |
underground | -0.026 |
Token | n |
Token ID | 299 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
jackets | +0.012 |
boats | +0.012 |
shells | +0.011 |
island | +0.011 |
rocks | +0.010 |
Neg logit contributions | |
istent | -0.015 |
oled | -0.014 |
urance | -0.014 |
opl | -0.014 |
Deal | -0.014 |
Token | uzz |
Token ID | 4715 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
arp | +0.027 |
island | +0.023 |
cubes | +0.023 |
bulbs | +0.021 |
place | +0.020 |
Neg logit contributions | |
grin | -0.039 |
thank | -0.039 |
amaz | -0.039 |
severe | -0.038 |
Deal | -0.038 |
Token | les |
Token ID | 829 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
arp | +0.019 |
cubes | +0.016 |
island | +0.016 |
bulbs | +0.015 |
boats | +0.014 |
Neg logit contributions | |
amaz | -0.032 |
Deal | -0.031 |
grin | -0.030 |
Once | -0.030 |
istent | -0.029 |
Token | them |
Token ID | 606 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
Deal | +0.053 |
istent | +0.051 |
amaz | +0.050 |
aman | +0.050 |
mmm | +0.049 |
Neg logit contributions | |
boats | -0.016 |
cubes | -0.013 |
creatures | -0.012 |
island | -0.012 |
shells | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
island | +0.032 |
cubes | +0.032 |
boats | +0.032 |
jackets | +0.030 |
tunnels | +0.028 |
Neg logit contributions | |
Deal | -0.065 |
aman | -0.063 |
istent | -0.063 |
amaz | -0.061 |
oled | -0.060 |
Token | He |
Token ID | 679 |
Feature activation | -8.14e-04 |
Pos logit contributions | |
arp | +0.015 |
cubes | +0.014 |
pipes | +0.013 |
bulbs | +0.013 |
island | +0.013 |
Neg logit contributions | |
used | -0.021 |
zing | -0.020 |
severe | -0.020 |
opl | -0.020 |
istent | -0.020 |
Token | had |
Token ID | 550 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
arp | +0.019 |
island | +0.017 |
cubes | +0.016 |
bulbs | +0.016 |
place | +0.016 |
Neg logit contributions | |
Deal | -0.030 |
grin | -0.029 |
amaz | -0.029 |
severe | -0.028 |
opl | -0.028 |
Token | lots |
Token ID | 6041 |
Feature activation | -5.43e-02 |
Pos logit contributions | |
island | +0.028 |
arp | +0.027 |
cubes | +0.026 |
bulbs | +0.026 |
pipes | +0.024 |
Neg logit contributions | |
Deal | -0.039 |
grin | -0.037 |
istent | -0.036 |
used | -0.036 |
ishly | -0.035 |
Token | of |
Token ID | 286 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
Deal | +0.114 |
severe | +0.105 |
ishly | +0.102 |
Jeffrey | +0.100 |
used | +0.099 |
Neg logit contributions | |
arp | -0.058 |
cubes | -0.057 |
place | -0.054 |
bulbs | -0.051 |
island | -0.050 |
Token | shiny |
Token ID | 22441 |
Feature activation | -5.07e-03 |
Pos logit contributions | |
severe | +0.037 |
Deal | +0.036 |
grin | +0.036 |
adays | +0.036 |
ishly | +0.035 |
Neg logit contributions | |
arp | -0.021 |
cubes | -0.020 |
island | -0.018 |
bulbs | -0.018 |
pipes | -0.017 |
Token | tables |
Token ID | 8893 |
Feature activation | -2.75e-02 |
Pos logit contributions | |
amaz | +0.009 |
grin | +0.009 |
Deal | +0.009 |
severe | +0.009 |
opl | +0.009 |
Neg logit contributions | |
arp | -0.006 |
island | -0.005 |
cubes | -0.005 |
bulbs | -0.005 |
place | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | -8.95e-03 |
Pos logit contributions | |
Deal | +0.062 |
severe | +0.054 |
Majesty | +0.054 |
grin | +0.052 |
amaz | +0.051 |
Neg logit contributions | |
island | -0.029 |
arp | -0.027 |
cubes | -0.027 |
underground | -0.023 |
bulbs | -0.023 |
Token | knew |
Token ID | 2993 |
Feature activation | +7.66e-03 |
Pos logit contributions | |
Deal | +0.020 |
severe | +0.019 |
grin | +0.019 |
istent | +0.019 |
opl | +0.019 |
Neg logit contributions | |
cubes | -0.008 |
arp | -0.007 |
island | -0.007 |
bulbs | -0.007 |
pipes | -0.006 |
Token | gadgets |
Token ID | 35281 |
Feature activation | -5.00e-03 |
Pos logit contributions | |
bulbs | +0.009 |
island | +0.008 |
shells | +0.008 |
crystals | +0.008 |
arp | +0.008 |
Neg logit contributions | |
opl | -0.015 |
Deal | -0.015 |
severe | -0.015 |
grin | -0.014 |
istent | -0.014 |
Token | everywhere |
Token ID | 8347 |
Feature activation | -3.61e-02 |
Pos logit contributions | |
Deal | +0.011 |
Aud | +0.010 |
amaz | +0.009 |
grin | +0.009 |
severe | +0.009 |
Neg logit contributions | |
island | -0.005 |
arp | -0.005 |
cubes | -0.005 |
bulbs | -0.005 |
place | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -9.07e-03 |
Pos logit contributions | |
Deal | +0.074 |
severe | +0.074 |
grin | +0.072 |
amaz | +0.071 |
opl | +0.071 |
Neg logit contributions | |
arp | -0.037 |
bulbs | -0.033 |
cubes | -0.032 |
island | -0.032 |
crystals | -0.028 |
Token | She |
Token ID | 1375 |
Feature activation | -5.25e-02 |
Pos logit contributions | |
severe | +0.010 |
opl | +0.010 |
grin | +0.009 |
istent | +0.009 |
Deal | +0.009 |
Neg logit contributions | |
arp | -0.016 |
island | -0.015 |
cubes | -0.015 |
bulbs | -0.014 |
collections | -0.013 |
Token | 'd |
Token ID | 1549 |
Feature activation | -5.74e-03 |
Pos logit contributions | |
Deal | +0.047 |
icks | +0.046 |
severe | +0.045 |
opl | +0.044 |
adays | +0.043 |
Neg logit contributions | |
island | -0.016 |
place | -0.014 |
arp | -0.013 |
ship | -0.010 |
tunnel | -0.010 |
Token | moved |
Token ID | 3888 |
Feature activation | +4.29e-03 |
Pos logit contributions | |
Deal | +0.010 |
severe | +0.010 |
ishly | +0.010 |
oled | +0.009 |
Immediately | +0.009 |
Neg logit contributions | |
island | -0.007 |
place | -0.006 |
arp | -0.006 |
tunnel | -0.005 |
treasures | -0.005 |
Token | far |
Token ID | 1290 |
Feature activation | -8.11e-02 |
Pos logit contributions | |
arp | +0.005 |
island | +0.005 |
cubes | +0.004 |
place | +0.004 |
bulbs | +0.004 |
Neg logit contributions | |
Deal | -0.008 |
opl | -0.008 |
oled | -0.008 |
istent | -0.007 |
adays | -0.007 |
Token | away |
Token ID | 1497 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
Deal | +0.156 |
opl | +0.147 |
Aud | +0.147 |
oled | +0.147 |
icks | +0.146 |
Neg logit contributions | |
arp | -0.091 |
island | -0.090 |
place | -0.074 |
cubes | -0.070 |
bulbs | -0.064 |
Token | from |
Token ID | 422 |
Feature activation | -2.85e-02 |
Pos logit contributions | |
Deal | +0.024 |
Aud | +0.022 |
opl | +0.021 |
oled | +0.021 |
icks | +0.020 |
Neg logit contributions | |
island | -0.011 |
creatures | -0.009 |
underground | -0.009 |
arp | -0.009 |
treasures | -0.008 |
Token | his |
Token ID | 465 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
oled | +0.051 |
istent | +0.050 |
Deal | +0.050 |
Aud | +0.050 |
ives | +0.048 |
Neg logit contributions | |
island | -0.038 |
cubes | -0.033 |
arp | -0.031 |
underground | -0.031 |
boats | -0.030 |
Token | old |
Token ID | 1468 |
Feature activation | -2.40e-02 |
Pos logit contributions | |
arp | +0.013 |
island | +0.013 |
cubes | +0.012 |
place | +0.011 |
ship | +0.010 |
Neg logit contributions | |
Deal | -0.020 |
oled | -0.020 |
icks | -0.020 |
adays | -0.020 |
severe | -0.020 |
Token | house |
Token ID | 2156 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
Deal | +0.048 |
adays | +0.045 |
amaz | +0.045 |
severe | +0.045 |
icks | +0.045 |
Neg logit contributions | |
arp | -0.025 |
island | -0.023 |
cubes | -0.022 |
bulbs | -0.020 |
pipes | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -6.24e-03 |
Pos logit contributions | |
Deal | +0.036 |
icks | +0.032 |
severe | +0.032 |
Aud | +0.031 |
opl | +0.031 |
Neg logit contributions | |
island | -0.017 |
arp | -0.014 |
cubes | -0.014 |
ship | -0.014 |
underground | -0.012 |
Token | With |
Token ID | 2080 |
Feature activation | -5.02e-02 |
Pos logit contributions | |
grin | +0.009 |
Aud | +0.009 |
icks | +0.009 |
la | +0.009 |
amaz | +0.009 |
Neg logit contributions | |
arp | -0.008 |
cubes | -0.007 |
leaning | -0.006 |
island | -0.006 |
ney | -0.006 |
Token | his |
Token ID | 465 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
oled | +0.093 |
istent | +0.088 |
AZ | +0.086 |
Deal | +0.086 |
Majesty | +0.085 |
Neg logit contributions | |
island | -0.056 |
cubes | -0.051 |
anchor | -0.048 |
minerals | -0.048 |
jackets | -0.048 |
Token | When |
Token ID | 1649 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
arp | +0.020 |
cubes | +0.019 |
pipes | +0.017 |
island | +0.017 |
Tim | +0.017 |
Neg logit contributions | |
grin | -0.026 |
opl | -0.026 |
Aud | -0.025 |
icks | -0.025 |
adays | -0.025 |
Token | he |
Token ID | 339 |
Feature activation | -2.57e-03 |
Pos logit contributions | |
Deal | +0.058 |
istent | +0.056 |
opl | +0.054 |
oled | +0.054 |
icks | +0.054 |
Neg logit contributions | |
island | -0.022 |
cubes | -0.017 |
ship | -0.016 |
ships | -0.016 |
boats | -0.015 |
Token | reached |
Token ID | 4251 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
Deal | +0.005 |
opl | +0.005 |
icks | +0.005 |
severe | +0.005 |
istent | +0.004 |
Neg logit contributions | |
island | -0.003 |
arp | -0.003 |
place | -0.002 |
ship | -0.002 |
cubes | -0.002 |
Token | a |
Token ID | 257 |
Feature activation | -7.89e-03 |
Pos logit contributions | |
island | +0.052 |
shells | +0.043 |
underground | +0.042 |
tunnels | +0.039 |
tunnel | +0.039 |
Neg logit contributions | |
istent | -0.087 |
oled | -0.085 |
Deal | -0.083 |
severe | -0.075 |
grin | -0.075 |
Token | clearing |
Token ID | 17304 |
Feature activation | +7.84e-03 |
Pos logit contributions | |
Deal | +0.015 |
icks | +0.014 |
severe | +0.014 |
opl | +0.014 |
istent | +0.014 |
Neg logit contributions | |
island | -0.010 |
arp | -0.009 |
tunnel | -0.008 |
place | -0.008 |
cubes | -0.008 |
Token | in |
Token ID | 287 |
Feature activation | +6.37e-02 |
Pos logit contributions | |
island | +0.012 |
arp | +0.011 |
cubes | +0.011 |
place | +0.010 |
bulbs | +0.010 |
Neg logit contributions | |
Deal | -0.013 |
severe | -0.012 |
Aud | -0.012 |
opl | -0.012 |
oled | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | +1.33e-03 |
Pos logit contributions | |
arp | +0.102 |
island | +0.102 |
cubes | +0.092 |
place | +0.088 |
bulbs | +0.088 |
Neg logit contributions | |
Deal | -0.085 |
severe | -0.084 |
istent | -0.083 |
oled | -0.083 |
opl | -0.081 |
Token | forest |
Token ID | 8222 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
arp | +0.002 |
island | +0.001 |
cubes | +0.001 |
place | +0.001 |
bulbs | +0.001 |
Neg logit contributions | |
grin | -0.002 |
amaz | -0.002 |
Deal | -0.002 |
severe | -0.002 |
opl | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | +1.08e-03 |
Pos logit contributions | |
island | +0.025 |
tunnel | +0.021 |
arp | +0.021 |
creatures | +0.020 |
place | +0.020 |
Neg logit contributions | |
Deal | -0.029 |
Aud | -0.026 |
severe | -0.025 |
opl | -0.024 |
icks | -0.023 |
Token | he |
Token ID | 339 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
island | +0.001 |
cubes | +0.001 |
pipes | +0.001 |
arp | +0.001 |
crystals | +0.001 |
Neg logit contributions | |
opl | -0.002 |
Deal | -0.002 |
istent | -0.002 |
icks | -0.002 |
severe | -0.002 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
opl | +0.031 |
Deal | +0.030 |
severe | +0.028 |
icks | +0.028 |
adays | +0.028 |
Neg logit contributions | |
island | -0.010 |
arp | -0.007 |
ship | -0.007 |
cubes | -0.006 |
ships | -0.006 |
Token | he |
Token ID | 339 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
Deal | +0.023 |
severe | +0.022 |
amaz | +0.022 |
opl | +0.022 |
istent | +0.022 |
Neg logit contributions | |
cubes | -0.013 |
island | -0.012 |
boats | -0.011 |
arp | -0.011 |
bulbs | -0.011 |
Token | snapped |
Token ID | 20821 |
Feature activation | -4.92e-03 |
Pos logit contributions | |
Deal | +0.038 |
amaz | +0.035 |
grin | +0.035 |
adays | +0.034 |
severe | +0.033 |
Neg logit contributions | |
arp | -0.020 |
island | -0.019 |
cubes | -0.017 |
place | -0.017 |
bulbs | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | -7.47e-03 |
Pos logit contributions | |
opl | +0.010 |
Deal | +0.010 |
istent | +0.009 |
Aud | +0.009 |
� | +0.009 |
Neg logit contributions | |
cubes | -0.006 |
crystals | -0.005 |
shells | -0.005 |
bulbs | -0.005 |
pipes | -0.005 |
Token | bark |
Token ID | 21405 |
Feature activation | -3.48e-02 |
Pos logit contributions | |
Deal | +0.016 |
istent | +0.015 |
oled | +0.014 |
Aud | +0.014 |
opl | +0.014 |
Neg logit contributions | |
cubes | -0.008 |
pipes | -0.007 |
tunnel | -0.007 |
crystals | -0.007 |
shells | -0.007 |
Token | ed |
Token ID | 276 |
Feature activation | -8.12e-02 |
Pos logit contributions | |
Deal | +0.061 |
amaz | +0.054 |
opl | +0.051 |
grin | +0.050 |
Jeffrey | +0.047 |
Neg logit contributions | |
arp | -0.053 |
cubes | -0.046 |
place | -0.045 |
B | -0.044 |
bulbs | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | -7.27e-03 |
Pos logit contributions | |
Deal | +0.183 |
opl | +0.167 |
Aud | +0.165 |
Sean | +0.162 |
� | +0.161 |
Neg logit contributions | |
cubes | -0.071 |
arp | -0.069 |
bulbs | -0.068 |
island | -0.068 |
creatures | -0.066 |
Token | One |
Token ID | 1881 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
amaz | +0.010 |
grin | +0.010 |
Aud | +0.010 |
severe | +0.009 |
adays | +0.009 |
Neg logit contributions | |
arp | -0.011 |
cubes | -0.011 |
island | -0.010 |
bulbs | -0.010 |
pipes | -0.009 |
Token | day |
Token ID | 1110 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
arp | +0.016 |
island | +0.013 |
cubes | +0.013 |
bulbs | +0.012 |
place | +0.011 |
Neg logit contributions | |
amaz | -0.031 |
grin | -0.031 |
opl | -0.030 |
Deal | -0.030 |
thank | -0.030 |
Token | his |
Token ID | 465 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
Deal | +0.026 |
opl | +0.025 |
amaz | +0.025 |
Aud | +0.024 |
grin | +0.023 |
Neg logit contributions | |
cubes | -0.031 |
arp | -0.030 |
island | -0.029 |
pipes | -0.027 |
bulbs | -0.027 |
Token | dad |
Token ID | 9955 |
Feature activation | +2.98e-03 |
Pos logit contributions | |
island | +0.011 |
arp | +0.011 |
place | +0.011 |
cubes | +0.009 |
tunnel | +0.009 |
Neg logit contributions | |
opl | -0.027 |
oled | -0.027 |
istent | -0.027 |
Immediately | -0.027 |
amaz | -0.026 |
Token | gave |
Token ID | 2921 |
Feature activation | -4.01e-02 |
Pos logit contributions | |
island | +0.003 |
arp | +0.003 |
cubes | +0.003 |
bulbs | +0.002 |
ship | +0.002 |
Neg logit contributions | |
Deal | -0.006 |
opl | -0.006 |
amaz | -0.006 |
grin | -0.006 |
severe | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
arp | +0.007 |
cubes | +0.007 |
island | +0.007 |
bulbs | +0.007 |
place | +0.006 |
Neg logit contributions | |
amaz | -0.003 |
opl | -0.003 |
grin | -0.003 |
Deal | -0.003 |
severe | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | +1.36e-03 |
Pos logit contributions | |
amaz | +0.008 |
grin | +0.007 |
thank | +0.007 |
Deal | +0.007 |
Immediately | +0.007 |
Neg logit contributions | |
arp | -0.027 |
island | -0.025 |
cubes | -0.024 |
place | -0.024 |
bulbs | -0.024 |
Token | lonely |
Token ID | 21757 |
Feature activation | +6.11e-03 |
Pos logit contributions | |
arp | +0.003 |
island | +0.002 |
bulbs | +0.002 |
cubes | +0.002 |
place | +0.002 |
Neg logit contributions | |
thank | -0.001 |
bye | -0.001 |
grin | -0.001 |
amaz | -0.001 |
Deal | -0.001 |
Token | bird |
Token ID | 6512 |
Feature activation | +2.90e-02 |
Pos logit contributions | |
arp | +0.009 |
island | +0.008 |
cubes | +0.007 |
bulbs | +0.007 |
place | +0.007 |
Neg logit contributions | |
Deal | -0.010 |
amaz | -0.009 |
severe | -0.009 |
grin | -0.009 |
opl | -0.009 |
Token | was |
Token ID | 373 |
Feature activation | -3.92e-03 |
Pos logit contributions | |
arp | +0.050 |
cubes | +0.046 |
island | +0.043 |
bulbs | +0.042 |
place | +0.042 |
Neg logit contributions | |
opl | -0.034 |
thank | -0.033 |
grin | -0.032 |
bye | -0.032 |
amaz | -0.031 |
Token | flying |
Token ID | 7348 |
Feature activation | -3.91e-02 |
Pos logit contributions | |
amaz | +0.006 |
Deal | +0.005 |
grin | +0.005 |
severe | +0.005 |
adays | +0.005 |
Neg logit contributions | |
arp | -0.006 |
island | -0.006 |
cubes | -0.006 |
bulbs | -0.005 |
place | -0.005 |
Token | in |
Token ID | 287 |
Feature activation | +7.56e-02 |
Pos logit contributions | |
grin | +0.057 |
amaz | +0.056 |
severe | +0.055 |
opl | +0.055 |
thank | +0.055 |
Neg logit contributions | |
arp | -0.057 |
island | -0.050 |
cubes | -0.049 |
bulbs | -0.046 |
place | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
island | +0.126 |
arp | +0.123 |
cubes | +0.117 |
bulbs | +0.113 |
tunnel | +0.107 |
Neg logit contributions | |
Deal | -0.098 |
oled | -0.097 |
grin | -0.097 |
severe | -0.096 |
istent | -0.096 |
Token | sky |
Token ID | 6766 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
arp | +0.017 |
island | +0.015 |
cubes | +0.015 |
bulbs | +0.014 |
place | +0.013 |
Neg logit contributions | |
amaz | -0.021 |
grin | -0.021 |
Deal | -0.020 |
severe | -0.020 |
opl | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
Deal | +0.027 |
amaz | +0.025 |
severe | +0.025 |
grin | +0.025 |
adays | +0.025 |
Neg logit contributions | |
arp | -0.020 |
island | -0.019 |
cubes | -0.018 |
place | -0.016 |
bulbs | -0.016 |
Token | The |
Token ID | 383 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
arp | +0.023 |
island | +0.022 |
cubes | +0.021 |
bulbs | +0.020 |
place | +0.020 |
Neg logit contributions | |
amaz | -0.009 |
Deal | -0.009 |
severe | -0.009 |
grin | -0.008 |
opl | -0.008 |
Token | Mom |
Token ID | 29252 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
arp | +0.082 |
cubes | +0.075 |
place | +0.063 |
island | +0.063 |
pipes | +0.059 |
Neg logit contributions | |
adays | -0.127 |
severe | -0.127 |
ishly | -0.126 |
amaz | -0.126 |
Deal | -0.126 |
Token | , |
Token ID | 11 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
cube | +0.033 |
arp | +0.030 |
cubes | +0.030 |
ship | +0.029 |
ie | +0.029 |
Neg logit contributions | |
amaz | -0.054 |
Aud | -0.053 |
Deal | -0.052 |
severe | -0.049 |
opl | -0.048 |
Token | the |
Token ID | 262 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
opl | +0.045 |
severe | +0.043 |
Aud | +0.042 |
Immediately | +0.041 |
Deal | +0.040 |
Neg logit contributions | |
cubes | -0.034 |
island | -0.032 |
arp | -0.032 |
cube | -0.031 |
anchor | -0.029 |
Token | bird |
Token ID | 6512 |
Feature activation | -3.29e-03 |
Pos logit contributions | |
oled | +0.022 |
severe | +0.022 |
adays | +0.022 |
Deal | +0.022 |
istent | +0.021 |
Neg logit contributions | |
island | -0.009 |
arp | -0.009 |
cubes | -0.008 |
bulbs | -0.007 |
place | -0.007 |
Token | ate |
Token ID | 15063 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
Deal | +0.007 |
amaz | +0.007 |
Aud | +0.006 |
severe | +0.006 |
grin | +0.006 |
Neg logit contributions | |
arp | -0.003 |
island | -0.003 |
cube | -0.003 |
ie | -0.003 |
bulb | -0.003 |
Token | my |
Token ID | 616 |
Feature activation | -7.13e-02 |
Pos logit contributions | |
oled | +0.036 |
grin | +0.035 |
Aud | +0.035 |
severe | +0.034 |
istent | +0.034 |
Neg logit contributions | |
cubes | -0.021 |
arp | -0.017 |
island | -0.017 |
crystals | -0.016 |
shells | -0.016 |
Token | fry |
Token ID | 32959 |
Feature activation | -8.33e-03 |
Pos logit contributions | |
adays | +0.127 |
severe | +0.127 |
oled | +0.127 |
Aud | +0.123 |
amaz | +0.119 |
Neg logit contributions | |
cubes | -0.085 |
arp | -0.083 |
island | -0.082 |
place | -0.073 |
cube | -0.072 |
Token | ! |
Token ID | 0 |
Feature activation | -6.72e-02 |
Pos logit contributions | |
amaz | +0.017 |
Deal | +0.016 |
Aud | +0.016 |
severe | +0.015 |
Majesty | +0.015 |
Neg logit contributions | |
cubes | -0.010 |
arp | -0.009 |
cube | -0.008 |
island | -0.008 |
place | -0.008 |
Token | â |
Token ID | 22940 |
Feature activation | +3.14e-02 |
Pos logit contributions | |
Aud | +0.103 |
severe | +0.099 |
zing | +0.097 |
grin | +0.096 |
hered | +0.095 |
Neg logit contributions | |
arp | -0.083 |
cubes | -0.077 |
Tim | -0.070 |
cube | -0.067 |
bulbs | -0.066 |
Token | € |
Token ID | 26391 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
arp | +0.054 |
island | +0.050 |
cubes | +0.049 |
bulbs | +0.047 |
place | +0.046 |
Neg logit contributions | |
grin | -0.037 |
amaz | -0.037 |
Deal | -0.036 |
opl | -0.036 |
severe | -0.036 |
Token |
|
Token ID | 198 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
arp | +0.022 |
cubes | +0.021 |
ney | +0.020 |
island | +0.019 |
collections | +0.018 |
Neg logit contributions | |
Deal | -0.022 |
severe | -0.021 |
amaz | -0.020 |
reflex | -0.020 |
grin | -0.019 |
Token | They |
Token ID | 1119 |
Feature activation | +1.00e-02 |
Pos logit contributions | |
arp | +0.008 |
island | +0.008 |
cubes | +0.008 |
bulbs | +0.007 |
place | +0.007 |
Neg logit contributions | |
grin | -0.006 |
Deal | -0.006 |
severe | -0.006 |
amaz | -0.006 |
Aud | -0.006 |
Token | chose |
Token ID | 7690 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
island | +0.010 |
arp | +0.009 |
bulbs | +0.009 |
cubes | +0.008 |
boats | +0.008 |
Neg logit contributions | |
Deal | -0.021 |
opl | -0.021 |
severe | -0.019 |
Aud | -0.018 |
grin | -0.018 |
Token | to |
Token ID | 284 |
Feature activation | -1.53e-03 |
Pos logit contributions | |
island | +0.015 |
bulbs | +0.014 |
cubes | +0.013 |
shells | +0.013 |
arp | +0.013 |
Neg logit contributions | |
Deal | -0.022 |
istent | -0.020 |
opl | -0.019 |
oled | -0.019 |
severe | -0.019 |
Token | explore |
Token ID | 7301 |
Feature activation | +4.32e-02 |
Pos logit contributions | |
oled | +0.003 |
Deal | +0.003 |
severe | +0.003 |
opl | +0.003 |
istent | +0.003 |
Neg logit contributions | |
island | -0.002 |
shells | -0.002 |
place | -0.002 |
bulbs | -0.001 |
boats | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | -1.24e-04 |
Pos logit contributions | |
island | +0.068 |
arp | +0.066 |
cubes | +0.061 |
bulbs | +0.061 |
place | +0.060 |
Neg logit contributions | |
Deal | -0.065 |
severe | -0.061 |
oled | -0.060 |
grin | -0.059 |
opl | -0.058 |
Token | land |
Token ID | 1956 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
amaz | +0.000 |
grin | +0.000 |
thank | +0.000 |
opl | +0.000 |
severe | +0.000 |
Neg logit contributions | |
arp | -0.000 |
cubes | -0.000 |
island | -0.000 |
bulbs | -0.000 |
place | -0.000 |
Token | together |
Token ID | 1978 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
Deal | +0.065 |
severe | +0.059 |
Aud | +0.057 |
grin | +0.055 |
oled | +0.055 |
Neg logit contributions | |
island | -0.046 |
arp | -0.041 |
cubes | -0.040 |
place | -0.037 |
ship | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | +6.22e-03 |
Pos logit contributions | |
island | +0.021 |
arp | +0.017 |
cubes | +0.016 |
place | +0.016 |
creatures | +0.015 |
Neg logit contributions | |
Deal | -0.042 |
severe | -0.040 |
Aud | -0.039 |
opl | -0.038 |
oled | -0.037 |
Token | They |
Token ID | 1119 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
arp | +0.011 |
island | +0.010 |
cubes | +0.010 |
bulbs | +0.009 |
place | +0.009 |
Neg logit contributions | |
grin | -0.007 |
severe | -0.007 |
Aud | -0.007 |
amaz | -0.007 |
Deal | -0.007 |
Token | wanted |
Token ID | 2227 |
Feature activation | -3.35e-03 |
Pos logit contributions | |
island | +0.013 |
arp | +0.012 |
bulbs | +0.011 |
cubes | +0.011 |
collections | +0.010 |
Neg logit contributions | |
Deal | -0.023 |
opl | -0.023 |
severe | -0.022 |
grin | -0.021 |
oled | -0.021 |
Token | to |
Token ID | 284 |
Feature activation | -4.38e-05 |
Pos logit contributions | |
Deal | +0.005 |
opl | +0.005 |
grin | +0.005 |
severe | +0.005 |
oled | +0.005 |
Neg logit contributions | |
arp | -0.004 |
island | -0.004 |
cubes | -0.004 |
place | -0.004 |
bulbs | -0.004 |
Token | pretty |
Token ID | 2495 |
Feature activation | +1.48e-03 |
Pos logit contributions | |
arp | +0.031 |
cubes | +0.029 |
bulbs | +0.027 |
place | +0.027 |
island | +0.026 |
Neg logit contributions | |
Deal | -0.029 |
amaz | -0.028 |
opl | -0.028 |
grin | -0.028 |
severe | -0.027 |
Token | her |
Token ID | 607 |
Feature activation | -1.46e-04 |
Pos logit contributions | |
bulbs | +0.002 |
cubes | +0.002 |
arp | +0.002 |
crystals | +0.002 |
jackets | +0.002 |
Neg logit contributions | |
Deal | -0.003 |
istent | -0.002 |
severe | -0.002 |
used | -0.002 |
Goodbye | -0.002 |
Token | hair |
Token ID | 4190 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
severe | +0.000 |
istent | +0.000 |
oled | +0.000 |
Jeffrey | +0.000 |
adays | +0.000 |
Neg logit contributions | |
cubes | -0.000 |
bulbs | -0.000 |
arp | -0.000 |
place | -0.000 |
jackets | -0.000 |
Token | looked |
Token ID | 3114 |
Feature activation | +3.83e-02 |
Pos logit contributions | |
Deal | +0.076 |
amaz | +0.067 |
Jeffrey | +0.067 |
severe | +0.066 |
used | +0.065 |
Neg logit contributions | |
arp | -0.025 |
cubes | -0.023 |
bulbs | -0.021 |
cube | -0.019 |
place | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -1.53e-04 |
Pos logit contributions | |
cubes | +0.033 |
arp | +0.032 |
island | +0.032 |
bulbs | +0.030 |
place | +0.029 |
Neg logit contributions | |
Deal | -0.081 |
Jeffrey | -0.079 |
Aud | -0.078 |
opl | -0.078 |
Josh | -0.077 |
Token | She |
Token ID | 1375 |
Feature activation | -4.06e-02 |
Pos logit contributions | |
opl | +0.000 |
used | +0.000 |
amaz | +0.000 |
grin | +0.000 |
severe | +0.000 |
Neg logit contributions | |
arp | -0.000 |
bulbs | -0.000 |
cubes | -0.000 |
island | -0.000 |
leaning | -0.000 |
Token | felt |
Token ID | 2936 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
Deal | +0.081 |
severe | +0.077 |
opl | +0.077 |
istent | +0.073 |
adays | +0.073 |
Neg logit contributions | |
arp | -0.037 |
cubes | -0.037 |
island | -0.037 |
bulbs | -0.033 |
cube | -0.031 |
Token | happy |
Token ID | 3772 |
Feature activation | -3.98e-04 |
Pos logit contributions | |
place | +0.015 |
arp | +0.014 |
cubes | +0.013 |
island | +0.013 |
bulbs | +0.013 |
Neg logit contributions | |
opl | -0.023 |
Deal | -0.023 |
istent | -0.022 |
used | -0.022 |
Majesty | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | +4.53e-04 |
Pos logit contributions | |
Deal | +0.001 |
opl | +0.001 |
severe | +0.001 |
Aud | +0.001 |
icks | +0.001 |
Neg logit contributions | |
island | -0.000 |
arp | -0.000 |
cubes | -0.000 |
bulbs | -0.000 |
crystals | -0.000 |
Token | proud |
Token ID | 6613 |
Feature activation | -5.14e-04 |
Pos logit contributions | |
arp | +0.000 |
island | +0.000 |
cubes | +0.000 |
organized | +0.000 |
creatures | +0.000 |
Neg logit contributions | |
Deal | -0.001 |
opl | -0.001 |
severe | -0.001 |
icks | -0.001 |
amaz | -0.001 |
Token | that |
Token ID | 326 |
Feature activation | -1.54e-02 |
Pos logit contributions | |
Deal | +0.001 |
opl | +0.001 |
severe | +0.001 |
grin | +0.001 |
amaz | +0.001 |
Neg logit contributions | |
arp | -0.001 |
cubes | -0.001 |
island | -0.001 |
bulbs | -0.001 |
place | -0.001 |
Token | store |
Token ID | 3650 |
Feature activation | -2.76e-02 |
Pos logit contributions | |
grin | +0.005 |
istent | +0.005 |
severe | +0.005 |
oled | +0.005 |
Deal | +0.005 |
Neg logit contributions | |
island | -0.003 |
arp | -0.003 |
cubes | -0.003 |
place | -0.003 |
bulbs | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | -7.86e-03 |
Pos logit contributions | |
grin | +0.046 |
Deal | +0.046 |
amaz | +0.046 |
severe | +0.045 |
oled | +0.045 |
Neg logit contributions | |
arp | -0.035 |
island | -0.034 |
cubes | -0.033 |
place | -0.031 |
bulbs | -0.029 |
Token | buy |
Token ID | 2822 |
Feature activation | -1.73e-03 |
Pos logit contributions | |
opl | +0.015 |
istent | +0.014 |
oled | +0.014 |
grin | +0.013 |
Deal | +0.013 |
Neg logit contributions | |
island | -0.010 |
cubes | -0.009 |
place | -0.009 |
bulbs | -0.009 |
arp | -0.008 |
Token | some |
Token ID | 617 |
Feature activation | +1.58e-03 |
Pos logit contributions | |
grin | +0.003 |
oled | +0.002 |
istent | +0.002 |
severe | +0.002 |
Deal | +0.002 |
Neg logit contributions | |
arp | -0.003 |
cubes | -0.003 |
island | -0.003 |
bulbs | -0.002 |
place | -0.002 |
Token | jam |
Token ID | 16853 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
arp | +0.002 |
island | +0.002 |
cubes | +0.002 |
bulbs | +0.002 |
place | +0.002 |
Neg logit contributions | |
grin | -0.002 |
amaz | -0.002 |
severe | -0.002 |
Deal | -0.002 |
oled | -0.002 |
Token | ? |
Token ID | 30 |
Feature activation | -3.48e-02 |
Pos logit contributions | |
amaz | +0.032 |
severe | +0.031 |
grin | +0.030 |
Deal | +0.030 |
oled | +0.029 |
Neg logit contributions | |
arp | -0.021 |
cubes | -0.021 |
island | -0.021 |
bulbs | -0.019 |
place | -0.019 |
Token | It |
Token ID | 632 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
grin | +0.051 |
opl | +0.050 |
amaz | +0.047 |
zing | +0.045 |
adays | +0.044 |
Neg logit contributions | |
arp | -0.053 |
bulbs | -0.047 |
cubes | -0.045 |
island | -0.044 |
leaning | -0.043 |
Token | is |
Token ID | 318 |
Feature activation | -5.56e-02 |
Pos logit contributions | |
arp | +0.024 |
island | +0.024 |
bulbs | +0.023 |
cubes | +0.022 |
place | +0.021 |
Neg logit contributions | |
grin | -0.034 |
oled | -0.034 |
amaz | -0.034 |
severe | -0.033 |
opl | -0.033 |
Token | not |
Token ID | 407 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
severe | +0.094 |
oled | +0.093 |
used | +0.092 |
grin | +0.091 |
istent | +0.089 |
Neg logit contributions | |
island | -0.071 |
cubes | -0.070 |
bulbs | -0.069 |
arp | -0.065 |
crystals | -0.062 |
Token | far |
Token ID | 1290 |
Feature activation | -7.73e-02 |
Pos logit contributions | |
amaz | +0.022 |
grin | +0.021 |
oled | +0.021 |
opl | +0.021 |
Deal | +0.021 |
Neg logit contributions | |
island | -0.019 |
cubes | -0.018 |
arp | -0.018 |
bulbs | -0.018 |
crystals | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -1.00e-02 |
Pos logit contributions | |
amaz | +0.139 |
opl | +0.134 |
Deal | +0.134 |
Aud | +0.132 |
grin | +0.129 |
Neg logit contributions | |
island | -0.092 |
arp | -0.089 |
cubes | -0.081 |
place | -0.078 |
bulbs | -0.076 |
Token | Wh |
Token ID | 854 |
Feature activation | -3.55e-02 |
Pos logit contributions | |
arp | +0.042 |
island | +0.039 |
cubes | +0.039 |
bulbs | +0.039 |
place | +0.036 |
Neg logit contributions | |
amaz | -0.027 |
grin | -0.026 |
severe | -0.025 |
opl | -0.024 |
Deal | -0.023 |
Token | isk |
Token ID | 1984 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
opl | +0.017 |
Deal | +0.016 |
grin | +0.016 |
amaz | +0.015 |
oled | +0.015 |
Neg logit contributions | |
arp | -0.085 |
island | -0.081 |
cubes | -0.080 |
bulbs | -0.077 |
place | -0.076 |
Token | ers |
Token ID | 364 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
opl | +0.280 |
Aud | +0.274 |
grin | +0.274 |
adays | +0.272 |
severe | +0.272 |
Neg logit contributions | |
arp | -0.229 |
island | -0.204 |
cubes | -0.198 |
place | -0.182 |
bulbs | -0.181 |
Token | . |
Token ID | 13 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
thank | +0.023 |
opl | +0.023 |
grin | +0.022 |
oled | +0.022 |
amaz | +0.021 |
Neg logit contributions | |
arp | -0.037 |
island | -0.034 |
cubes | -0.033 |
bulbs | -0.032 |
place | -0.032 |
Token | Wh |
Token ID | 854 |
Feature activation | -3.58e-02 |
Pos logit contributions | |
amaz | +0.014 |
grin | +0.014 |
Deal | +0.014 |
severe | +0.013 |
opl | +0.013 |
Neg logit contributions | |
arp | -0.027 |
island | -0.025 |
cubes | -0.025 |
bulbs | -0.024 |
place | -0.023 |
Token | isk |
Token ID | 1984 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
opl | +0.015 |
Deal | +0.013 |
oled | +0.012 |
grin | +0.012 |
severe | +0.011 |
Neg logit contributions | |
arp | -0.089 |
island | -0.085 |
cubes | -0.083 |
bulbs | -0.080 |
place | -0.079 |
Token | ers |
Token ID | 364 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
opl | +0.254 |
Aud | +0.247 |
icks | +0.241 |
adays | +0.241 |
severe | +0.239 |
Neg logit contributions | |
arp | -0.250 |
island | -0.226 |
cubes | -0.219 |
place | -0.203 |
bulbs | -0.200 |
Token | liked |
Token ID | 8288 |
Feature activation | -7.09e-03 |
Pos logit contributions | |
thank | +0.019 |
grin | +0.017 |
Not | +0.016 |
opl | +0.015 |
bye | +0.015 |
Neg logit contributions | |
arp | -0.043 |
isher | -0.036 |
cubes | -0.035 |
island | -0.035 |
bulbs | -0.035 |
Token | to |
Token ID | 284 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
amaz | +0.009 |
grin | +0.009 |
Deal | +0.009 |
severe | +0.009 |
opl | +0.009 |
Neg logit contributions | |
arp | -0.011 |
island | -0.010 |
cubes | -0.010 |
bulbs | -0.010 |
place | -0.010 |
Token | imagine |
Token ID | 5967 |
Feature activation | -3.89e-03 |
Pos logit contributions | |
amaz | +0.029 |
oled | +0.028 |
used | +0.028 |
Deal | +0.028 |
severe | +0.028 |
Neg logit contributions | |
island | -0.021 |
arp | -0.021 |
cubes | -0.019 |
bulbs | -0.019 |
place | -0.019 |
Token | that |
Token ID | 326 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
Deal | +0.006 |
severe | +0.006 |
grin | +0.006 |
amaz | +0.006 |
oled | +0.006 |
Neg logit contributions | |
arp | -0.005 |
island | -0.005 |
cubes | -0.005 |
bulbs | -0.005 |
place | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
Deal | +0.131 |
severe | +0.122 |
icks | +0.120 |
upid | +0.115 |
Aud | +0.114 |
Neg logit contributions | |
island | -0.044 |
arp | -0.035 |
cubes | -0.032 |
occasion | -0.029 |
place | -0.029 |
Token | still |
Token ID | 991 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
icks | +0.035 |
Deal | +0.034 |
opl | +0.033 |
aud | +0.032 |
severe | +0.032 |
Neg logit contributions | |
island | -0.016 |
cubes | -0.013 |
ship | -0.013 |
ships | -0.012 |
boy | -0.012 |
Token | in |
Token ID | 287 |
Feature activation | +4.98e-02 |
Pos logit contributions | |
Deal | +0.032 |
Aud | +0.030 |
severe | +0.029 |
oled | +0.029 |
icks | +0.028 |
Neg logit contributions | |
island | -0.019 |
creatures | -0.016 |
place | -0.016 |
underground | -0.015 |
cubes | -0.015 |
Token | his |
Token ID | 465 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
island | +0.060 |
cubes | +0.055 |
place | +0.053 |
place | +0.052 |
tunnel | +0.052 |
Neg logit contributions | |
elcome | -0.085 |
Aud | -0.084 |
oled | -0.084 |
severe | -0.083 |
istent | -0.082 |
Token | p |
Token ID | 279 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
island | +0.013 |
place | +0.011 |
cubes | +0.010 |
collections | +0.010 |
arp | +0.009 |
Neg logit contributions | |
oled | -0.025 |
istent | -0.023 |
severe | -0.023 |
adays | -0.021 |
Aud | -0.021 |
Token | aj |
Token ID | 1228 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
arp | +0.013 |
cubes | +0.011 |
island | +0.009 |
place | +0.009 |
bulbs | +0.009 |
Neg logit contributions | |
severe | -0.019 |
thank | -0.019 |
Deal | -0.019 |
adays | -0.019 |
Jeffrey | -0.019 |
Token | amas |
Token ID | 17485 |
Feature activation | -8.68e-02 |
Pos logit contributions | |
thank | +0.218 |
catches | +0.215 |
severe | +0.214 |
icks | +0.212 |
Deal | +0.207 |
Neg logit contributions | |
arp | -0.213 |
cubes | -0.193 |
island | -0.185 |
bulbs | -0.167 |
pans | -0.162 |
Token | . |
Token ID | 13 |
Feature activation | -9.37e-03 |
Pos logit contributions | |
Deal | +0.215 |
Majesty | +0.189 |
icks | +0.188 |
amaz | +0.187 |
Aud | +0.186 |
Neg logit contributions | |
island | -0.064 |
arp | -0.062 |
washer | -0.053 |
ship | -0.050 |
cubes | -0.050 |
Token | Everyone |
Token ID | 11075 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
Aud | +0.016 |
icks | +0.016 |
severe | +0.015 |
hered | +0.015 |
grin | +0.014 |
Neg logit contributions | |
arp | -0.010 |
cubes | -0.009 |
ney | -0.008 |
garage | -0.008 |
island | -0.008 |
Token | in |
Token ID | 287 |
Feature activation | +4.37e-02 |
Pos logit contributions | |
Deal | +0.018 |
opl | +0.018 |
severe | +0.018 |
amaz | +0.018 |
grin | +0.017 |
Neg logit contributions | |
arp | -0.014 |
island | -0.013 |
cubes | -0.012 |
place | -0.012 |
bulbs | -0.011 |
Token | town |
Token ID | 3240 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
island | +0.049 |
cubes | +0.042 |
boats | +0.040 |
ships | +0.039 |
ship | +0.038 |
Neg logit contributions | |
severe | -0.079 |
Deal | -0.078 |
istent | -0.078 |
opl | -0.077 |
elcome | -0.077 |
Token | small |
Token ID | 1402 |
Feature activation | -2.19e-02 |
Pos logit contributions | |
grin | +0.071 |
amaz | +0.071 |
Deal | +0.070 |
severe | +0.069 |
opl | +0.067 |
Neg logit contributions | |
arp | -0.099 |
island | -0.094 |
cubes | -0.090 |
bulbs | -0.086 |
place | -0.084 |
Token | , |
Token ID | 11 |
Feature activation | -9.61e-03 |
Pos logit contributions | |
Deal | +0.039 |
Aud | +0.038 |
amaz | +0.036 |
grin | +0.036 |
oled | +0.035 |
Neg logit contributions | |
island | -0.030 |
arp | -0.027 |
place | -0.027 |
cubes | -0.027 |
bulbs | -0.026 |
Token | has |
Token ID | 468 |
Feature activation | -3.52e-02 |
Pos logit contributions | |
severe | +0.016 |
oled | +0.016 |
Aud | +0.016 |
opl | +0.016 |
used | +0.015 |
Neg logit contributions | |
arp | -0.013 |
island | -0.013 |
cubes | -0.012 |
bulbs | -0.012 |
cube | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
grin | +0.058 |
Deal | +0.055 |
oled | +0.054 |
istent | +0.053 |
Aud | +0.050 |
Neg logit contributions | |
island | -0.056 |
cubes | -0.053 |
arp | -0.051 |
bulbs | -0.050 |
pipes | -0.049 |
Token | bus |
Token ID | 1323 |
Feature activation | -7.30e-02 |
Pos logit contributions | |
adays | +0.050 |
severe | +0.050 |
alled | +0.050 |
Aud | +0.049 |
oled | +0.049 |
Neg logit contributions | |
island | -0.032 |
pipe | -0.029 |
arp | -0.028 |
tunnel | -0.028 |
cell | -0.028 |
Token | hy |
Token ID | 12114 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
Deal | +0.142 |
amaz | +0.138 |
delight | +0.134 |
fascination | +0.132 |
ease | +0.132 |
Neg logit contributions | |
island | -0.078 |
ship | -0.071 |
arp | -0.068 |
place | -0.067 |
cubes | -0.066 |
Token | tail |
Token ID | 7894 |
Feature activation | -8.11e-02 |
Pos logit contributions | |
Deal | +0.262 |
� | +0.254 |
amaz | +0.252 |
Afterwards | +0.252 |
adays | +0.250 |
Neg logit contributions | |
island | -0.188 |
ship | -0.172 |
cubes | -0.155 |
arp | -0.153 |
an | -0.153 |
Token | , |
Token ID | 11 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
Deal | +0.178 |
amaz | +0.177 |
Aud | +0.169 |
grin | +0.163 |
adays | +0.162 |
Neg logit contributions | |
island | -0.079 |
arp | -0.069 |
cubes | -0.062 |
ship | -0.059 |
cube | -0.055 |
Token | and |
Token ID | 290 |
Feature activation | -7.17e-03 |
Pos logit contributions | |
Deal | +0.026 |
severe | +0.026 |
hered | +0.025 |
oled | +0.025 |
Aud | +0.025 |
Neg logit contributions | |
island | -0.017 |
pipes | -0.016 |
cubes | -0.016 |
bulbs | -0.016 |
coal | -0.015 |
Token | likes |
Token ID | 7832 |
Feature activation | -6.39e-02 |
Pos logit contributions | |
oled | +0.014 |
grin | +0.013 |
Deal | +0.013 |
istent | +0.013 |
Aud | +0.013 |
Neg logit contributions | |
island | -0.009 |
bulbs | -0.009 |
crystals | -0.009 |
pipes | -0.009 |
cubes | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | -9.43e-03 |
Pos logit contributions | |
grin | +0.125 |
Deal | +0.123 |
amaz | +0.117 |
oled | +0.116 |
istent | +0.113 |
Neg logit contributions | |
island | -0.077 |
arp | -0.073 |
cubes | -0.069 |
bulbs | -0.068 |
place | -0.066 |
Token | a |
Token ID | 257 |
Feature activation | -2.91e-02 |
Pos logit contributions | |
amaz | +0.021 |
grin | +0.020 |
Deal | +0.020 |
thank | +0.019 |
opl | +0.018 |
Neg logit contributions | |
arp | -0.068 |
island | -0.063 |
cubes | -0.062 |
place | -0.060 |
bulbs | -0.059 |
Token | friendly |
Token ID | 8030 |
Feature activation | -2.88e-02 |
Pos logit contributions | |
thank | +0.028 |
grin | +0.027 |
bye | +0.026 |
amaz | +0.025 |
sorry | +0.024 |
Neg logit contributions | |
arp | -0.059 |
cubes | -0.051 |
island | -0.051 |
place | -0.050 |
bulbs | -0.050 |
Token | cat |
Token ID | 3797 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
Deal | +0.049 |
amaz | +0.047 |
severe | +0.047 |
opl | +0.046 |
adays | +0.046 |
Neg logit contributions | |
arp | -0.037 |
island | -0.036 |
cubes | -0.033 |
bulbs | -0.031 |
place | -0.029 |
Token | named |
Token ID | 3706 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
opl | +0.023 |
thank | +0.022 |
icks | +0.021 |
oled | +0.021 |
Mitt | +0.020 |
Neg logit contributions | |
arp | -0.038 |
cubes | -0.034 |
island | -0.034 |
place | -0.033 |
collections | -0.032 |
Token | Wh |
Token ID | 854 |
Feature activation | -3.55e-02 |
Pos logit contributions | |
arp | +0.042 |
island | +0.039 |
cubes | +0.039 |
bulbs | +0.039 |
place | +0.036 |
Neg logit contributions | |
amaz | -0.027 |
grin | -0.026 |
severe | -0.025 |
opl | -0.024 |
Deal | -0.023 |
Token | isk |
Token ID | 1984 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
opl | +0.017 |
Deal | +0.016 |
grin | +0.016 |
amaz | +0.015 |
oled | +0.015 |
Neg logit contributions | |
arp | -0.085 |
island | -0.081 |
cubes | -0.080 |
bulbs | -0.077 |
place | -0.076 |
Token | ers |
Token ID | 364 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
opl | +0.280 |
Aud | +0.274 |
grin | +0.274 |
adays | +0.272 |
severe | +0.272 |
Neg logit contributions | |
arp | -0.229 |
island | -0.204 |
cubes | -0.198 |
place | -0.182 |
bulbs | -0.181 |
Token | . |
Token ID | 13 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
thank | +0.023 |
opl | +0.023 |
grin | +0.022 |
oled | +0.022 |
amaz | +0.021 |
Neg logit contributions | |
arp | -0.037 |
island | -0.034 |
cubes | -0.033 |
bulbs | -0.032 |
place | -0.032 |
Token | Wh |
Token ID | 854 |
Feature activation | -3.58e-02 |
Pos logit contributions | |
amaz | +0.014 |
grin | +0.014 |
Deal | +0.014 |
severe | +0.013 |
opl | +0.013 |
Neg logit contributions | |
arp | -0.027 |
island | -0.025 |
cubes | -0.025 |
bulbs | -0.024 |
place | -0.023 |
Token | isk |
Token ID | 1984 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
opl | +0.015 |
Deal | +0.013 |
oled | +0.012 |
grin | +0.012 |
severe | +0.011 |
Neg logit contributions | |
arp | -0.089 |
island | -0.085 |
cubes | -0.083 |
bulbs | -0.080 |
place | -0.079 |
Token | ers |
Token ID | 364 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
opl | +0.254 |
Aud | +0.247 |
icks | +0.241 |
adays | +0.241 |
severe | +0.239 |
Neg logit contributions | |
arp | -0.250 |
island | -0.226 |
cubes | -0.219 |
place | -0.203 |
bulbs | -0.200 |
Token | drawer |
Token ID | 33451 |
Feature activation | +1.35e-02 |
Pos logit contributions | |
amaz | +0.004 |
grin | +0.004 |
Deal | +0.004 |
severe | +0.004 |
opl | +0.004 |
Neg logit contributions | |
arp | -0.005 |
island | -0.004 |
cubes | -0.004 |
bulbs | -0.004 |
place | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | -8.06e-03 |
Pos logit contributions | |
arp | +0.021 |
island | +0.020 |
cubes | +0.019 |
place | +0.018 |
bulbs | +0.018 |
Neg logit contributions | |
Deal | -0.019 |
amaz | -0.019 |
grin | -0.019 |
opl | -0.018 |
severe | -0.018 |
Token | get |
Token ID | 651 |
Feature activation | -3.96e-02 |
Pos logit contributions | |
amaz | +0.010 |
grin | +0.010 |
Deal | +0.010 |
thank | +0.010 |
severe | +0.009 |
Neg logit contributions | |
arp | -0.014 |
island | -0.012 |
cubes | -0.012 |
bulbs | -0.012 |
place | -0.011 |
Token | her |
Token ID | 607 |
Feature activation | -2.83e-03 |
Pos logit contributions | |
amaz | +0.026 |
thank | +0.026 |
grin | +0.024 |
Deal | +0.023 |
severe | +0.023 |
Neg logit contributions | |
arp | -0.091 |
island | -0.083 |
cubes | -0.078 |
place | -0.077 |
leaning | -0.075 |
Token | cr |
Token ID | 1067 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
amaz | +0.005 |
grin | +0.005 |
Deal | +0.005 |
severe | +0.005 |
opl | +0.005 |
Neg logit contributions | |
arp | -0.004 |
island | -0.003 |
cubes | -0.003 |
bulbs | -0.003 |
place | -0.003 |
Token | ay |
Token ID | 323 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
amaz | +0.007 |
grin | +0.007 |
Deal | +0.007 |
severe | +0.006 |
thank | +0.006 |
Neg logit contributions | |
arp | -0.027 |
island | -0.025 |
cubes | -0.025 |
bulbs | -0.024 |
place | -0.024 |
Token | ons |
Token ID | 684 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
amaz | +0.151 |
grin | +0.150 |
Deal | +0.147 |
severe | +0.143 |
opl | +0.141 |
Neg logit contributions | |
arp | -0.238 |
island | -0.217 |
cubes | -0.215 |
bulbs | -0.204 |
place | -0.201 |
Token | , |
Token ID | 11 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
amaz | +0.024 |
Deal | +0.024 |
grin | +0.024 |
opl | +0.023 |
severe | +0.023 |
Neg logit contributions | |
arp | -0.021 |
island | -0.020 |
cubes | -0.019 |
bulbs | -0.018 |
place | -0.017 |
Token | but |
Token ID | 475 |
Feature activation | +1.40e-02 |
Pos logit contributions | |
amaz | +0.028 |
grin | +0.028 |
opl | +0.028 |
Deal | +0.028 |
severe | +0.028 |
Neg logit contributions | |
arp | -0.021 |
island | -0.019 |
cubes | -0.019 |
bulbs | -0.018 |
place | -0.017 |
Token | she |
Token ID | 673 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
arp | +0.021 |
island | +0.021 |
cubes | +0.020 |
bulbs | +0.020 |
crystals | +0.018 |
Neg logit contributions | |
opl | -0.020 |
grin | -0.020 |
Deal | -0.020 |
istent | -0.019 |
severe | -0.019 |
Token | saw |
Token ID | 2497 |
Feature activation | +6.85e-03 |
Pos logit contributions | |
amaz | +0.025 |
grin | +0.025 |
Deal | +0.024 |
severe | +0.024 |
opl | +0.024 |
Neg logit contributions | |
arp | -0.035 |
island | -0.032 |
cubes | -0.031 |
bulbs | -0.030 |
place | -0.029 |