Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 2868 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.5388. Min Range: -0.5388

Max Act: 0.4777. Min Act: -0.0001

Data Index: 191256 (C4 (Web Text))

Max Activating Token Index: 631

Text #1

Max Range: 0.5388. Min Range: -0.5388

Max Act: 0.5388. Min Act: -0.0001

Data Index: 1283226 (C4 (Web Text))

Max Activating Token Index: 697

Text #2

Max Range: 0.5388. Min Range: -0.5388

Max Act: 0.4745. Min Act: -0.0001

Data Index: 217983 (C4 (Web Text))

Max Activating Token Index: 43

Text #3

Max Range: 0.5388. Min Range: -0.5388

Max Act: 0.4877. Min Act: -0.0001

Data Index: 898664 (C4 (Web Text))

Max Activating Token Index: 745

Text #4

Max Range: 0.5388. Min Range: -0.5388

Max Act: 0.4804. Min Act: -0.0001

Data Index: 483793 (C4 (Web Text))

Max Activating Token Index: 58

Text #5

Max Range: 0.5388. Min Range: -0.5388

Max Act: 0.4482. Min Act: -0.0001

Data Index: 1046424 (C4 (Web Text))

Max Activating Token Index: 780

Text #6

Max Range: 0.5388. Min Range: -0.5388

Max Act: 0.4577. Min Act: -0.0001

Data Index: 882443 (C4 (Web Text))

Max Activating Token Index: 702

Text #7

Max Range: 0.5388. Min Range: -0.5388

Max Act: 0.4497. Min Act: -0.0001

Data Index: 82112 (C4 (Web Text))

Max Activating Token Index: 780

Text #8

Max Range: 0.5388. Min Range: -0.5388

Max Act: 0.4469. Min Act: -0.0001

Data Index: 193588 (C4 (Web Text))

Max Activating Token Index: 895

Text #9

Max Range: 0.5388. Min Range: -0.5388

Max Act: 0.4468. Min Act: -0.0001

Data Index: 586258 (C4 (Web Text))

Max Activating Token Index: 62

Text #10

Max Range: 0.5388. Min Range: -0.5388

Max Act: 0.4415. Min Act: -0.0001

Data Index: 424425 (C4 (Web Text))

Max Activating Token Index: 75

Text #11

Max Range: 0.5388. Min Range: -0.5388

Max Act: 0.3949. Min Act: -0.0001

Data Index: 1344000 (C4 (Web Text))

Max Activating Token Index: 640

Text #12

Max Range: 0.5388. Min Range: -0.5388

Max Act: 0.4191. Min Act: -0.0001

Data Index: 1266097 (C4 (Web Text))

Max Activating Token Index: 489

Text #13

Max Range: 0.5388. Min Range: -0.5388

Max Act: 0.4054. Min Act: -0.0001

Data Index: 484955 (C4 (Web Text))

Max Activating Token Index: 1001

Text #14

Max Range: 0.5388. Min Range: -0.5388

Max Act: 0.3989. Min Act: -0.0001

Data Index: 231492 (C4 (Web Text))

Max Activating Token Index: 947

Text #15

Max Range: 0.5388. Min Range: -0.5388

Max Act: 0.4116. Min Act: -0.0001

Data Index: 695377 (C4 (Web Text))

Max Activating Token Index: 905

Text #16

Max Range: 0.5388. Min Range: -0.5388

Max Act: 0.4168. Min Act: -0.0001

Data Index: 453251 (C4 (Web Text))

Max Activating Token Index: 371

Text #17

Max Range: 0.5388. Min Range: -0.5388

Max Act: 0.4008. Min Act: -0.0001

Data Index: 122506 (C4 (Web Text))

Max Activating Token Index: 188

Text #18

Max Range: 0.5388. Min Range: -0.5388

Max Act: 0.4028. Min Act: -0.0001

Data Index: 782516 (C4 (Web Text))

Max Activating Token Index: 83

Text #19

Max Range: 0.5388. Min Range: -0.5388

Max Act: 0.3946. Min Act: -0.0001

Data Index: 130950 (C4 (Web Text))

Max Activating Token Index: 565