Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 3145 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.4532. Min Range: -0.4532

Max Act: 0.4532. Min Act: -0.0000

Data Index: 305530 (C4 (Web Text))

Max Activating Token Index: 518

Text #1

Max Range: 0.4532. Min Range: -0.4532

Max Act: 0.4276. Min Act: -0.0000

Data Index: 537305 (C4 (Web Text))

Max Activating Token Index: 746

Text #2

Max Range: 0.4532. Min Range: -0.4532

Max Act: 0.3854. Min Act: -0.0000

Data Index: 1085688 (C4 (Web Text))

Max Activating Token Index: 684

Text #3

Max Range: 0.4532. Min Range: -0.4532

Max Act: 0.3771. Min Act: -0.0001

Data Index: 860289 (C4 (Web Text))

Max Activating Token Index: 952

Text #4

Max Range: 0.4532. Min Range: -0.4532

Max Act: 0.3795. Min Act: -0.0000

Data Index: 153712 (C4 (Web Text))

Max Activating Token Index: 950

Text #5

Max Range: 0.4532. Min Range: -0.4532

Max Act: 0.3828. Min Act: -0.0000

Data Index: 974355 (C4 (Web Text))

Max Activating Token Index: 396

Text #6

Max Range: 0.4532. Min Range: -0.4532

Max Act: 0.3656. Min Act: -0.0000

Data Index: 1197646 (C4 (Web Text))

Max Activating Token Index: 395

Text #7

Max Range: 0.4532. Min Range: -0.4532

Max Act: 0.3674. Min Act: -0.0000

Data Index: 238323 (C4 (Web Text))

Max Activating Token Index: 918

Text #8

Max Range: 0.4532. Min Range: -0.4532

Max Act: 0.3547. Min Act: -0.0001

Data Index: 714815 (C4 (Web Text))

Max Activating Token Index: 550

Text #9

Max Range: 0.4532. Min Range: -0.4532

Max Act: 0.3322. Min Act: -0.0001

Data Index: 1007153 (C4 (Web Text))

Max Activating Token Index: 1008

Text #10

Max Range: 0.4532. Min Range: -0.4532

Max Act: 0.3314. Min Act: -0.0000

Data Index: 328498 (C4 (Web Text))

Max Activating Token Index: 166

Text #11

Max Range: 0.4532. Min Range: -0.4532

Max Act: 0.3114. Min Act: -0.0001

Data Index: 1066488 (C4 (Web Text))

Max Activating Token Index: 804

Text #12

Max Range: 0.4532. Min Range: -0.4532

Max Act: 0.3229. Min Act: -0.0001

Data Index: 1690 (C4 (Web Text))

Max Activating Token Index: 601

Text #13

Max Range: 0.4532. Min Range: -0.4532

Max Act: 0.3101. Min Act: -0.0000

Data Index: 1006149 (C4 (Web Text))

Max Activating Token Index: 388

Text #14

Max Range: 0.4532. Min Range: -0.4532

Max Act: 0.3046. Min Act: -0.0000

Data Index: 748335 (C4 (Web Text))

Max Activating Token Index: 731

Text #15

Max Range: 0.4532. Min Range: -0.4532

Max Act: 0.3204. Min Act: -0.0000

Data Index: 541279 (C4 (Web Text))

Max Activating Token Index: 913

Text #16

Max Range: 0.4532. Min Range: -0.4532

Max Act: 0.2989. Min Act: -0.0000

Data Index: 42750 (C4 (Web Text))

Max Activating Token Index: 855

Text #17

Max Range: 0.4532. Min Range: -0.4532

Max Act: 0.3163. Min Act: -0.0000

Data Index: 1055352 (C4 (Web Text))

Max Activating Token Index: 772

Text #18

Max Range: 0.4532. Min Range: -0.4532

Max Act: 0.2899. Min Act: -0.0000

Data Index: 1198999 (C4 (Web Text))

Max Activating Token Index: 699

Text #19

Max Range: 0.4532. Min Range: -0.4532

Max Act: 0.2985. Min Act: -0.0000

Data Index: 182653 (C4 (Web Text))

Max Activating Token Index: 673