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

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

Neuron 217 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: 0.9251. Min Range: -0.9251

Max Act: 0.9251. Min Act: -0.0001

Data Index: 463799 (C4 (Web Text))

Max Activating Token Index: 535

Text #1

Max Range: 0.9251. Min Range: -0.9251

Max Act: 0.7937. Min Act: -0.0001

Data Index: 363215 (C4 (Web Text))

Max Activating Token Index: 539

Text #2

Max Range: 0.9251. Min Range: -0.9251

Max Act: 0.8239. Min Act: -0.0001

Data Index: 1106312 (C4 (Web Text))

Max Activating Token Index: 82

Text #3

Max Range: 0.9251. Min Range: -0.9251

Max Act: 0.8780. Min Act: -0.0001

Data Index: 901836 (C4 (Web Text))

Max Activating Token Index: 696

Text #4

Max Range: 0.9251. Min Range: -0.9251

Max Act: 0.7923. Min Act: -0.0001

Data Index: 580388 (C4 (Web Text))

Max Activating Token Index: 296

Text #5

Max Range: 0.9251. Min Range: -0.9251

Max Act: 0.8258. Min Act: -0.0001

Data Index: 757253 (C4 (Web Text))

Max Activating Token Index: 94

Text #6

Max Range: 0.9251. Min Range: -0.9251

Max Act: 0.8193. Min Act: -0.0001

Data Index: 266561 (C4 (Web Text))

Max Activating Token Index: 377

Text #7

Max Range: 0.9251. Min Range: -0.9251

Max Act: 0.8169. Min Act: -0.0001

Data Index: 285287 (C4 (Web Text))

Max Activating Token Index: 329

Text #8

Max Range: 0.9251. Min Range: -0.9251

Max Act: 0.7838. Min Act: -0.0001

Data Index: 455239 (C4 (Web Text))

Max Activating Token Index: 59

Text #9

Max Range: 0.9251. Min Range: -0.9251

Max Act: 0.7762. Min Act: -0.0001

Data Index: 961301 (C4 (Web Text))

Max Activating Token Index: 387

Text #10

Max Range: 0.9251. Min Range: -0.9251

Max Act: 0.7671. Min Act: -0.0001

Data Index: 933467 (C4 (Web Text))

Max Activating Token Index: 799

Text #11

Max Range: 0.9251. Min Range: -0.9251

Max Act: 0.8295. Min Act: -0.0001

Data Index: 1270483 (C4 (Web Text))

Max Activating Token Index: 897

Text #12

Max Range: 0.9251. Min Range: -0.9251

Max Act: 0.7199. Min Act: -0.0001

Data Index: 469330 (C4 (Web Text))

Max Activating Token Index: 808

Text #13

Max Range: 0.9251. Min Range: -0.9251

Max Act: 0.7432. Min Act: -0.0001

Data Index: 953953 (C4 (Web Text))

Max Activating Token Index: 998

Text #14

Max Range: 0.9251. Min Range: -0.9251

Max Act: 0.7416. Min Act: -0.0001

Data Index: 440697 (C4 (Web Text))

Max Activating Token Index: 74

Text #15

Max Range: 0.9251. Min Range: -0.9251

Max Act: 0.7910. Min Act: -0.0001

Data Index: 390906 (C4 (Web Text))

Max Activating Token Index: 299

Text #16

Max Range: 0.9251. Min Range: -0.9251

Max Act: 0.7192. Min Act: -0.0001

Data Index: 657620 (C4 (Web Text))

Max Activating Token Index: 151

Text #17

Max Range: 0.9251. Min Range: -0.9251

Max Act: 0.6786. Min Act: -0.0001

Data Index: 175551 (C4 (Web Text))

Max Activating Token Index: 625

Text #18

Max Range: 0.9251. Min Range: -0.9251

Max Act: 0.7168. Min Act: -0.0001

Data Index: 1136606 (C4 (Web Text))

Max Activating Token Index: 869

Text #19

Max Range: 0.9251. Min Range: -0.9251

Max Act: 0.7382. Min Act: -0.0001

Data Index: 521374 (C4 (Web Text))

Max Activating Token Index: 776