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

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

Neuron 3049 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.8375. Min Range: -4.8375

Max Act: 4.8233. Min Act: -0.0000

Data Index: 1220629 (C4 (Web Text))

Max Activating Token Index: 38

Text #1

Max Range: 4.8375. Min Range: -4.8375

Max Act: 4.8295. Min Act: -0.0000

Data Index: 740567 (C4 (Web Text))

Max Activating Token Index: 98

Text #2

Max Range: 4.8375. Min Range: -4.8375

Max Act: 4.8375. Min Act: -0.0000

Data Index: 1290402 (C4 (Web Text))

Max Activating Token Index: 116

Text #3

Max Range: 4.8375. Min Range: -4.8375

Max Act: 4.4168. Min Act: -0.0001

Data Index: 1273126 (C4 (Web Text))

Max Activating Token Index: 395

Text #4

Max Range: 4.8375. Min Range: -4.8375

Max Act: 4.5152. Min Act: -0.0000

Data Index: 41294 (C4 (Web Text))

Max Activating Token Index: 87

Text #5

Max Range: 4.8375. Min Range: -4.8375

Max Act: 4.1928. Min Act: -0.0000

Data Index: 648208 (C4 (Web Text))

Max Activating Token Index: 213

Text #6

Max Range: 4.8375. Min Range: -4.8375

Max Act: 4.3076. Min Act: -0.0000

Data Index: 1243028 (C4 (Web Text))

Max Activating Token Index: 101

Text #7

Max Range: 4.8375. Min Range: -4.8375

Max Act: 4.2799. Min Act: -0.0000

Data Index: 638242 (C4 (Web Text))

Max Activating Token Index: 427

Text #8

Max Range: 4.8375. Min Range: -4.8375

Max Act: 4.2811. Min Act: -0.0000

Data Index: 1221048 (C4 (Web Text))

Max Activating Token Index: 66

Text #9

Max Range: 4.8375. Min Range: -4.8375

Max Act: 3.9521. Min Act: -0.0000

Data Index: 593463 (C4 (Web Text))

Max Activating Token Index: 225

Text #10

Max Range: 4.8375. Min Range: -4.8375

Max Act: 4.2078. Min Act: -0.0000

Data Index: 764776 (C4 (Web Text))

Max Activating Token Index: 530

Text #11

Max Range: 4.8375. Min Range: -4.8375

Max Act: 4.3580. Min Act: -0.0000

Data Index: 1200202 (C4 (Web Text))

Max Activating Token Index: 464

Text #12

Max Range: 4.8375. Min Range: -4.8375

Max Act: 4.1952. Min Act: -0.0000

Data Index: 69389 (C4 (Web Text))

Max Activating Token Index: 118

Text #13

Max Range: 4.8375. Min Range: -4.8375

Max Act: 4.0176. Min Act: -0.0000

Data Index: 2315 (C4 (Web Text))

Max Activating Token Index: 142

Text #14

Max Range: 4.8375. Min Range: -4.8375

Max Act: 4.4329. Min Act: -0.0000

Data Index: 154036 (C4 (Web Text))

Max Activating Token Index: 439

Text #15

Max Range: 4.8375. Min Range: -4.8375

Max Act: 4.0647. Min Act: -0.0000

Data Index: 571786 (C4 (Web Text))

Max Activating Token Index: 1006

Text #16

Max Range: 4.8375. Min Range: -4.8375

Max Act: 4.0086. Min Act: -0.0000

Data Index: 837490 (C4 (Web Text))

Max Activating Token Index: 318

Text #17

Max Range: 4.8375. Min Range: -4.8375

Max Act: 3.7104. Min Act: -0.0000

Data Index: 924092 (C4 (Web Text))

Max Activating Token Index: 515

Text #18

Max Range: 4.8375. Min Range: -4.8375

Max Act: 4.1759. Min Act: -0.0000

Data Index: 609417 (C4 (Web Text))

Max Activating Token Index: 271

Text #19

Max Range: 4.8375. Min Range: -4.8375

Max Act: 3.9966. Min Act: -0.0000

Data Index: 220727 (C4 (Web Text))

Max Activating Token Index: 922