Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1351 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 5.3318. Min Range: -5.3318

Max Act: 5.3318. Min Act: -0.1700

Data Index: 807742 (C4 (Web Text))

Max Activating Token Index: 602

Text #1

Max Range: 5.3318. Min Range: -5.3318

Max Act: 5.2693. Min Act: -0.1700

Data Index: 540920 (C4 (Web Text))

Max Activating Token Index: 702

Text #2

Max Range: 5.3318. Min Range: -5.3318

Max Act: 5.1755. Min Act: -0.1700

Data Index: 455466 (C4 (Web Text))

Max Activating Token Index: 277

Text #3

Max Range: 5.3318. Min Range: -5.3318

Max Act: 5.1755. Min Act: -0.1669

Data Index: 417337 (C4 (Web Text))

Max Activating Token Index: 81

Text #4

Max Range: 5.3318. Min Range: -5.3318

Max Act: 5.2068. Min Act: -0.1700

Data Index: 1334438 (C4 (Web Text))

Max Activating Token Index: 240

Text #5

Max Range: 5.3318. Min Range: -5.3318

Max Act: 5.1130. Min Act: -0.1700

Data Index: 1246267 (C4 (Web Text))

Max Activating Token Index: 264

Text #6

Max Range: 5.3318. Min Range: -5.3318

Max Act: 5.1130. Min Act: -0.0860

Data Index: 1628849 (Python Code)

Max Activating Token Index: 376

Text #7

Max Range: 5.3318. Min Range: -5.3318

Max Act: 5.0818. Min Act: -0.1700

Data Index: 54316 (C4 (Web Text))

Max Activating Token Index: 512

Text #8

Max Range: 5.3318. Min Range: -5.3318

Max Act: 5.1130. Min Act: -0.1699

Data Index: 520163 (C4 (Web Text))

Max Activating Token Index: 572

Text #9

Max Range: 5.3318. Min Range: -5.3318

Max Act: 5.0505. Min Act: -0.1700

Data Index: 1342265 (C4 (Web Text))

Max Activating Token Index: 301

Text #10

Max Range: 5.3318. Min Range: -5.3318

Max Act: 5.0193. Min Act: -0.1700

Data Index: 255065 (C4 (Web Text))

Max Activating Token Index: 1012

Text #11

Max Range: 5.3318. Min Range: -5.3318

Max Act: 5.0505. Min Act: -0.1700

Data Index: 603387 (C4 (Web Text))

Max Activating Token Index: 363

Text #12

Max Range: 5.3318. Min Range: -5.3318

Max Act: 5.0818. Min Act: -0.1700

Data Index: 353851 (C4 (Web Text))

Max Activating Token Index: 588

Text #13

Max Range: 5.3318. Min Range: -5.3318

Max Act: 4.9880. Min Act: -0.1699

Data Index: 488759 (C4 (Web Text))

Max Activating Token Index: 470

Text #14

Max Range: 5.3318. Min Range: -5.3318

Max Act: 5.0193. Min Act: -0.1700

Data Index: 1341102 (C4 (Web Text))

Max Activating Token Index: 665

Text #15

Max Range: 5.3318. Min Range: -5.3318

Max Act: 5.0505. Min Act: -0.1700

Data Index: 540355 (C4 (Web Text))

Max Activating Token Index: 110

Text #16

Max Range: 5.3318. Min Range: -5.3318

Max Act: 5.0193. Min Act: -0.1700

Data Index: 1033265 (C4 (Web Text))

Max Activating Token Index: 738

Text #17

Max Range: 5.3318. Min Range: -5.3318

Max Act: 4.9880. Min Act: -0.1700

Data Index: 1350485 (C4 (Web Text))

Max Activating Token Index: 423

Text #18

Max Range: 5.3318. Min Range: -5.3318

Max Act: 5.0193. Min Act: -0.1700

Data Index: 18280 (C4 (Web Text))

Max Activating Token Index: 720

Text #19

Max Range: 5.3318. Min Range: -5.3318

Max Act: 4.9880. Min Act: -0.1697

Data Index: 1629208 (Python Code)

Max Activating Token Index: 981