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

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

Neuron 2197 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 1.0365. Min Range: -1.0365

Max Act: 1.0365. Min Act: -0.0001

Data Index: 1324711 (C4 (Web Text))

Max Activating Token Index: 757

Text #1

Max Range: 1.0365. Min Range: -1.0365

Max Act: 0.9525. Min Act: -0.0001

Data Index: 355618 (C4 (Web Text))

Max Activating Token Index: 240

Text #2

Max Range: 1.0365. Min Range: -1.0365

Max Act: 1.0199. Min Act: -0.0001

Data Index: 94137 (C4 (Web Text))

Max Activating Token Index: 927

Text #3

Max Range: 1.0365. Min Range: -1.0365

Max Act: 0.9010. Min Act: -0.0001

Data Index: 1082429 (C4 (Web Text))

Max Activating Token Index: 214

Text #4

Max Range: 1.0365. Min Range: -1.0365

Max Act: 0.8656. Min Act: -0.0001

Data Index: 16441 (C4 (Web Text))

Max Activating Token Index: 390

Text #5

Max Range: 1.0365. Min Range: -1.0365

Max Act: 0.8077. Min Act: -0.0001

Data Index: 260219 (C4 (Web Text))

Max Activating Token Index: 570

Text #6

Max Range: 1.0365. Min Range: -1.0365

Max Act: 0.8149. Min Act: -0.0001

Data Index: 456473 (C4 (Web Text))

Max Activating Token Index: 565

Text #7

Max Range: 1.0365. Min Range: -1.0365

Max Act: 0.7836. Min Act: -0.0001

Data Index: 603946 (C4 (Web Text))

Max Activating Token Index: 81

Text #8

Max Range: 1.0365. Min Range: -1.0365

Max Act: 0.7855. Min Act: -0.0001

Data Index: 728366 (C4 (Web Text))

Max Activating Token Index: 711

Text #9

Max Range: 1.0365. Min Range: -1.0365

Max Act: 0.7812. Min Act: -0.0001

Data Index: 879133 (C4 (Web Text))

Max Activating Token Index: 797

Text #10

Max Range: 1.0365. Min Range: -1.0365

Max Act: 0.8497. Min Act: -0.0001

Data Index: 192808 (C4 (Web Text))

Max Activating Token Index: 925

Text #11

Max Range: 1.0365. Min Range: -1.0365

Max Act: 0.8009. Min Act: -0.0001

Data Index: 395828 (C4 (Web Text))

Max Activating Token Index: 899

Text #12

Max Range: 1.0365. Min Range: -1.0365

Max Act: 0.6982. Min Act: -0.0001

Data Index: 923653 (C4 (Web Text))

Max Activating Token Index: 856

Text #13

Max Range: 1.0365. Min Range: -1.0365

Max Act: 0.7292. Min Act: -0.0001

Data Index: 1004383 (C4 (Web Text))

Max Activating Token Index: 649

Text #14

Max Range: 1.0365. Min Range: -1.0365

Max Act: 0.7067. Min Act: -0.0001

Data Index: 82667 (C4 (Web Text))

Max Activating Token Index: 897

Text #15

Max Range: 1.0365. Min Range: -1.0365

Max Act: 0.7456. Min Act: -0.0001

Data Index: 431784 (C4 (Web Text))

Max Activating Token Index: 339

Text #16

Max Range: 1.0365. Min Range: -1.0365

Max Act: 0.6511. Min Act: -0.0001

Data Index: 946123 (C4 (Web Text))

Max Activating Token Index: 186

Text #17

Max Range: 1.0365. Min Range: -1.0365

Max Act: 0.6744. Min Act: -0.0001

Data Index: 1489913 (Python Code)

Max Activating Token Index: 205

Text #18

Max Range: 1.0365. Min Range: -1.0365

Max Act: 0.6431. Min Act: -0.0001

Data Index: 1298348 (C4 (Web Text))

Max Activating Token Index: 22

Text #19

Max Range: 1.0365. Min Range: -1.0365

Max Act: 0.6249. Min Act: -0.0001

Data Index: 211130 (C4 (Web Text))

Max Activating Token Index: 315