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

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

Neuron 5634 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: 0.1288. Min Range: -0.1288

Max Act: 0.1236. Min Act: -0.0001

Data Index: 836417 (C4 (Web Text))

Max Activating Token Index: 431

Text #1

Max Range: 0.1288. Min Range: -0.1288

Max Act: 0.1288. Min Act: -0.0001

Data Index: 860757 (C4 (Web Text))

Max Activating Token Index: 729

Text #2

Max Range: 0.1288. Min Range: -0.1288

Max Act: 0.1278. Min Act: -0.0000

Data Index: 420041 (C4 (Web Text))

Max Activating Token Index: 805

Text #3

Max Range: 0.1288. Min Range: -0.1288

Max Act: 0.1162. Min Act: -0.0000

Data Index: 1140141 (C4 (Web Text))

Max Activating Token Index: 289

Text #4

Max Range: 0.1288. Min Range: -0.1288

Max Act: 0.1241. Min Act: -0.0001

Data Index: 737235 (C4 (Web Text))

Max Activating Token Index: 721

Text #5

Max Range: 0.1288. Min Range: -0.1288

Max Act: 0.1264. Min Act: -0.0000

Data Index: 1144478 (C4 (Web Text))

Max Activating Token Index: 546

Text #6

Max Range: 0.1288. Min Range: -0.1288

Max Act: 0.1156. Min Act: -0.0000

Data Index: 1048625 (C4 (Web Text))

Max Activating Token Index: 523

Text #7

Max Range: 0.1288. Min Range: -0.1288

Max Act: 0.1108. Min Act: -0.0001

Data Index: 984841 (C4 (Web Text))

Max Activating Token Index: 943

Text #8

Max Range: 0.1288. Min Range: -0.1288

Max Act: 0.1149. Min Act: -0.0000

Data Index: 805088 (C4 (Web Text))

Max Activating Token Index: 762

Text #9

Max Range: 0.1288. Min Range: -0.1288

Max Act: 0.1119. Min Act: -0.0000

Data Index: 1632946 (Python Code)

Max Activating Token Index: 442

Text #10

Max Range: 0.1288. Min Range: -0.1288

Max Act: 0.1248. Min Act: -0.0000

Data Index: 708711 (C4 (Web Text))

Max Activating Token Index: 956

Text #11

Max Range: 0.1288. Min Range: -0.1288

Max Act: 0.1182. Min Act: -0.0000

Data Index: 570610 (C4 (Web Text))

Max Activating Token Index: 541

Text #12

Max Range: 0.1288. Min Range: -0.1288

Max Act: 0.1107. Min Act: -0.0001

Data Index: 960680 (C4 (Web Text))

Max Activating Token Index: 63

Text #13

Max Range: 0.1288. Min Range: -0.1288

Max Act: 0.1089. Min Act: -0.0001

Data Index: 320757 (C4 (Web Text))

Max Activating Token Index: 253

Text #14

Max Range: 0.1288. Min Range: -0.1288

Max Act: 0.1200. Min Act: -0.0000

Data Index: 1424466 (Python Code)

Max Activating Token Index: 537

Text #15

Max Range: 0.1288. Min Range: -0.1288

Max Act: 0.0983. Min Act: -0.0000

Data Index: 574842 (C4 (Web Text))

Max Activating Token Index: 438

Text #16

Max Range: 0.1288. Min Range: -0.1288

Max Act: 0.1084. Min Act: -0.0000

Data Index: 581332 (C4 (Web Text))

Max Activating Token Index: 151

Text #17

Max Range: 0.1288. Min Range: -0.1288

Max Act: 0.1152. Min Act: -0.0000

Data Index: 151801 (C4 (Web Text))

Max Activating Token Index: 64

Text #18

Max Range: 0.1288. Min Range: -0.1288

Max Act: 0.1115. Min Act: -0.0000

Data Index: 571895 (C4 (Web Text))

Max Activating Token Index: 22

Text #19

Max Range: 0.1288. Min Range: -0.1288

Max Act: 0.1055. Min Act: -0.0000

Data Index: 1337680 (C4 (Web Text))

Max Activating Token Index: 390