Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 2550 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.5228. Min Range: -0.5228

Max Act: 0.5015. Min Act: -0.0001

Data Index: 1164389 (C4 (Web Text))

Max Activating Token Index: 875

Text #1

Max Range: 0.5228. Min Range: -0.5228

Max Act: 0.5228. Min Act: -0.0001

Data Index: 1329387 (C4 (Web Text))

Max Activating Token Index: 600

Text #2

Max Range: 0.5228. Min Range: -0.5228

Max Act: 0.5142. Min Act: -0.0001

Data Index: 375867 (C4 (Web Text))

Max Activating Token Index: 828

Text #3

Max Range: 0.5228. Min Range: -0.5228

Max Act: 0.4847. Min Act: -0.0001

Data Index: 882741 (C4 (Web Text))

Max Activating Token Index: 700

Text #4

Max Range: 0.5228. Min Range: -0.5228

Max Act: 0.4469. Min Act: -0.0001

Data Index: 896184 (C4 (Web Text))

Max Activating Token Index: 666

Text #5

Max Range: 0.5228. Min Range: -0.5228

Max Act: 0.4493. Min Act: -0.0001

Data Index: 1149531 (C4 (Web Text))

Max Activating Token Index: 367

Text #6

Max Range: 0.5228. Min Range: -0.5228

Max Act: 0.4599. Min Act: -0.0001

Data Index: 1005423 (C4 (Web Text))

Max Activating Token Index: 387

Text #7

Max Range: 0.5228. Min Range: -0.5228

Max Act: 0.4264. Min Act: -0.0001

Data Index: 954937 (C4 (Web Text))

Max Activating Token Index: 148

Text #8

Max Range: 0.5228. Min Range: -0.5228

Max Act: 0.4524. Min Act: -0.0001

Data Index: 775067 (C4 (Web Text))

Max Activating Token Index: 904

Text #9

Max Range: 0.5228. Min Range: -0.5228

Max Act: 0.4320. Min Act: -0.0001

Data Index: 385543 (C4 (Web Text))

Max Activating Token Index: 423

Text #10

Max Range: 0.5228. Min Range: -0.5228

Max Act: 0.4309. Min Act: -0.0001

Data Index: 770716 (C4 (Web Text))

Max Activating Token Index: 853

Text #11

Max Range: 0.5228. Min Range: -0.5228

Max Act: 0.4292. Min Act: -0.0001

Data Index: 1066446 (C4 (Web Text))

Max Activating Token Index: 891

Text #12

Max Range: 0.5228. Min Range: -0.5228

Max Act: 0.4252. Min Act: -0.0001

Data Index: 836322 (C4 (Web Text))

Max Activating Token Index: 186

Text #13

Max Range: 0.5228. Min Range: -0.5228

Max Act: 0.4173. Min Act: -0.0001

Data Index: 447850 (C4 (Web Text))

Max Activating Token Index: 35

Text #14

Max Range: 0.5228. Min Range: -0.5228

Max Act: 0.4299. Min Act: -0.0001

Data Index: 1105065 (C4 (Web Text))

Max Activating Token Index: 84

Text #15

Max Range: 0.5228. Min Range: -0.5228

Max Act: 0.4151. Min Act: -0.0001

Data Index: 1178215 (C4 (Web Text))

Max Activating Token Index: 41

Text #16

Max Range: 0.5228. Min Range: -0.5228

Max Act: 0.4140. Min Act: -0.0001

Data Index: 425005 (C4 (Web Text))

Max Activating Token Index: 395

Text #17

Max Range: 0.5228. Min Range: -0.5228

Max Act: 0.4122. Min Act: -0.0001

Data Index: 253165 (C4 (Web Text))

Max Activating Token Index: 686

Text #18

Max Range: 0.5228. Min Range: -0.5228

Max Act: 0.4118. Min Act: -0.0001

Data Index: 1056618 (C4 (Web Text))

Max Activating Token Index: 752

Text #19

Max Range: 0.5228. Min Range: -0.5228

Max Act: 0.4139. Min Act: -0.0001

Data Index: 1105090 (C4 (Web Text))

Max Activating Token Index: 423