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

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

Neuron 2576 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4162. Min Range: -0.4162

Max Act: 0.4162. Min Act: -0.0001

Data Index: 607214 (C4 (Web Text))

Max Activating Token Index: 235

Text #1

Max Range: 0.4162. Min Range: -0.4162

Max Act: 0.4054. Min Act: -0.0000

Data Index: 568507 (C4 (Web Text))

Max Activating Token Index: 857

Text #2

Max Range: 0.4162. Min Range: -0.4162

Max Act: 0.3603. Min Act: -0.0001

Data Index: 1178066 (C4 (Web Text))

Max Activating Token Index: 373

Text #3

Max Range: 0.4162. Min Range: -0.4162

Max Act: 0.3582. Min Act: -0.0001

Data Index: 53510 (C4 (Web Text))

Max Activating Token Index: 841

Text #4

Max Range: 0.4162. Min Range: -0.4162

Max Act: 0.3544. Min Act: -0.0001

Data Index: 194088 (C4 (Web Text))

Max Activating Token Index: 531

Text #5

Max Range: 0.4162. Min Range: -0.4162

Max Act: 0.3509. Min Act: -0.0001

Data Index: 866865 (C4 (Web Text))

Max Activating Token Index: 578

Text #6

Max Range: 0.4162. Min Range: -0.4162

Max Act: 0.3383. Min Act: -0.0000

Data Index: 806123 (C4 (Web Text))

Max Activating Token Index: 651

Text #7

Max Range: 0.4162. Min Range: -0.4162

Max Act: 0.3289. Min Act: -0.0001

Data Index: 1005492 (C4 (Web Text))

Max Activating Token Index: 381

Text #8

Max Range: 0.4162. Min Range: -0.4162

Max Act: 0.3562. Min Act: -0.0001

Data Index: 275999 (C4 (Web Text))

Max Activating Token Index: 537

Text #9

Max Range: 0.4162. Min Range: -0.4162

Max Act: 0.3557. Min Act: -0.0001

Data Index: 1387987 (Python Code)

Max Activating Token Index: 14

Text #10

Max Range: 0.4162. Min Range: -0.4162

Max Act: 0.3314. Min Act: -0.0001

Data Index: 796683 (C4 (Web Text))

Max Activating Token Index: 352

Text #11

Max Range: 0.4162. Min Range: -0.4162

Max Act: 0.3084. Min Act: -0.0000

Data Index: 86440 (C4 (Web Text))

Max Activating Token Index: 690

Text #12

Max Range: 0.4162. Min Range: -0.4162

Max Act: 0.3171. Min Act: -0.0001

Data Index: 318626 (C4 (Web Text))

Max Activating Token Index: 342

Text #13

Max Range: 0.4162. Min Range: -0.4162

Max Act: 0.3158. Min Act: -0.0001

Data Index: 550505 (C4 (Web Text))

Max Activating Token Index: 789

Text #14

Max Range: 0.4162. Min Range: -0.4162

Max Act: 0.3226. Min Act: -0.0001

Data Index: 72190 (C4 (Web Text))

Max Activating Token Index: 592

Text #15

Max Range: 0.4162. Min Range: -0.4162

Max Act: 0.3125. Min Act: -0.0001

Data Index: 495985 (C4 (Web Text))

Max Activating Token Index: 997

Text #16

Max Range: 0.4162. Min Range: -0.4162

Max Act: 0.3066. Min Act: -0.0001

Data Index: 807470 (C4 (Web Text))

Max Activating Token Index: 294

Text #17

Max Range: 0.4162. Min Range: -0.4162

Max Act: 0.3099. Min Act: -0.0001

Data Index: 751306 (C4 (Web Text))

Max Activating Token Index: 673

Text #18

Max Range: 0.4162. Min Range: -0.4162

Max Act: 0.2954. Min Act: -0.0001

Data Index: 675568 (C4 (Web Text))

Max Activating Token Index: 466

Text #19

Max Range: 0.4162. Min Range: -0.4162

Max Act: 0.3186. Min Act: -0.0001

Data Index: 151512 (C4 (Web Text))

Max Activating Token Index: 933