Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 3226 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Hooked Transformer Loading: HookedTransformer.from_pretrained('solu-12l')

Click the toggle to see the full text!

Text #0

Max Range: 0.8007. Min Range: -0.8007

Max Act: 0.8007. Min Act: -0.0002

Data Index: 1151554 (C4 (Web Text))

Max Activating Token Index: 434

Text #1

Max Range: 0.8007. Min Range: -0.8007

Max Act: 0.6342. Min Act: -0.0002

Data Index: 710332 (C4 (Web Text))

Max Activating Token Index: 443

Text #2

Max Range: 0.8007. Min Range: -0.8007

Max Act: 0.5029. Min Act: -0.0002

Data Index: 174227 (C4 (Web Text))

Max Activating Token Index: 907

Text #3

Max Range: 0.8007. Min Range: -0.8007

Max Act: 0.4445. Min Act: -0.0002

Data Index: 1073630 (C4 (Web Text))

Max Activating Token Index: 611

Text #4

Max Range: 0.8007. Min Range: -0.8007

Max Act: 0.4423. Min Act: -0.0002

Data Index: 282496 (C4 (Web Text))

Max Activating Token Index: 720

Text #5

Max Range: 0.8007. Min Range: -0.8007

Max Act: 0.4311. Min Act: -0.0002

Data Index: 963095 (C4 (Web Text))

Max Activating Token Index: 538

Text #6

Max Range: 0.8007. Min Range: -0.8007

Max Act: 0.4305. Min Act: -0.0002

Data Index: 4079 (C4 (Web Text))

Max Activating Token Index: 280

Text #7

Max Range: 0.8007. Min Range: -0.8007

Max Act: 0.4147. Min Act: -0.0002

Data Index: 779815 (C4 (Web Text))

Max Activating Token Index: 803

Text #8

Max Range: 0.8007. Min Range: -0.8007

Max Act: 0.4422. Min Act: -0.0002

Data Index: 1467216 (Python Code)

Max Activating Token Index: 659

Text #9

Max Range: 0.8007. Min Range: -0.8007

Max Act: 0.4386. Min Act: -0.0002

Data Index: 667855 (C4 (Web Text))

Max Activating Token Index: 30

Text #10

Max Range: 0.8007. Min Range: -0.8007

Max Act: 0.4258. Min Act: -0.0002

Data Index: 1126337 (C4 (Web Text))

Max Activating Token Index: 217

Text #11

Max Range: 0.8007. Min Range: -0.8007

Max Act: 0.4057. Min Act: -0.0002

Data Index: 301499 (C4 (Web Text))

Max Activating Token Index: 826

Text #12

Max Range: 0.8007. Min Range: -0.8007

Max Act: 0.4178. Min Act: -0.0002

Data Index: 546740 (C4 (Web Text))

Max Activating Token Index: 534

Text #13

Max Range: 0.8007. Min Range: -0.8007

Max Act: 0.4104. Min Act: -0.0003

Data Index: 67196 (C4 (Web Text))

Max Activating Token Index: 764

Text #14

Max Range: 0.8007. Min Range: -0.8007

Max Act: 0.4065. Min Act: -0.0002

Data Index: 597972 (C4 (Web Text))

Max Activating Token Index: 515

Text #15

Max Range: 0.8007. Min Range: -0.8007

Max Act: 0.4178. Min Act: -0.0002

Data Index: 1242336 (C4 (Web Text))

Max Activating Token Index: 756

Text #16

Max Range: 0.8007. Min Range: -0.8007

Max Act: 0.4061. Min Act: -0.0002

Data Index: 1333930 (C4 (Web Text))

Max Activating Token Index: 147

Text #17

Max Range: 0.8007. Min Range: -0.8007

Max Act: 0.4002. Min Act: -0.0002

Data Index: 241411 (C4 (Web Text))

Max Activating Token Index: 209

Text #18

Max Range: 0.8007. Min Range: -0.8007

Max Act: 0.4168. Min Act: -0.0002

Data Index: 87674 (C4 (Web Text))

Max Activating Token Index: 260

Text #19

Max Range: 0.8007. Min Range: -0.8007

Max Act: 0.4006. Min Act: -0.0002

Data Index: 926491 (C4 (Web Text))

Max Activating Token Index: 139