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

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

Neuron 1421 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: 5.3085. Min Range: -5.3085

Max Act: 5.3085. Min Act: -0.1700

Data Index: 1573426 (Python Code)

Max Activating Token Index: 781

Text #1

Max Range: 5.3085. Min Range: -5.3085

Max Act: 5.2929. Min Act: -0.1700

Data Index: 558976 (C4 (Web Text))

Max Activating Token Index: 212

Text #2

Max Range: 5.3085. Min Range: -5.3085

Max Act: 5.2148. Min Act: -0.1700

Data Index: 704430 (C4 (Web Text))

Max Activating Token Index: 160

Text #3

Max Range: 5.3085. Min Range: -5.3085

Max Act: 5.1992. Min Act: -0.1700

Data Index: 1019119 (C4 (Web Text))

Max Activating Token Index: 820

Text #4

Max Range: 5.3085. Min Range: -5.3085

Max Act: 5.1210. Min Act: -0.1700

Data Index: 701671 (C4 (Web Text))

Max Activating Token Index: 176

Text #5

Max Range: 5.3085. Min Range: -5.3085

Max Act: 5.0742. Min Act: -0.1700

Data Index: 1514383 (Python Code)

Max Activating Token Index: 221

Text #6

Max Range: 5.3085. Min Range: -5.3085

Max Act: 5.0429. Min Act: -0.1700

Data Index: 889239 (C4 (Web Text))

Max Activating Token Index: 622

Text #7

Max Range: 5.3085. Min Range: -5.3085

Max Act: 5.0195. Min Act: -0.1700

Data Index: 1163116 (C4 (Web Text))

Max Activating Token Index: 354

Text #8

Max Range: 5.3085. Min Range: -5.3085

Max Act: 5.0039. Min Act: -0.1700

Data Index: 516889 (C4 (Web Text))

Max Activating Token Index: 497

Text #9

Max Range: 5.3085. Min Range: -5.3085

Max Act: 5.0039. Min Act: -0.1700

Data Index: 116429 (C4 (Web Text))

Max Activating Token Index: 725

Text #10

Max Range: 5.3085. Min Range: -5.3085

Max Act: 4.9648. Min Act: -0.1700

Data Index: 764057 (C4 (Web Text))

Max Activating Token Index: 235

Text #11

Max Range: 5.3085. Min Range: -5.3085

Max Act: 4.9726. Min Act: -0.1700

Data Index: 1640709 (Python Code)

Max Activating Token Index: 108

Text #12

Max Range: 5.3085. Min Range: -5.3085

Max Act: 4.8710. Min Act: -0.1700

Data Index: 53610 (C4 (Web Text))

Max Activating Token Index: 991

Text #13

Max Range: 5.3085. Min Range: -5.3085

Max Act: 4.9023. Min Act: -0.1700

Data Index: 647276 (C4 (Web Text))

Max Activating Token Index: 521

Text #14

Max Range: 5.3085. Min Range: -5.3085

Max Act: 4.9335. Min Act: -0.1700

Data Index: 304710 (C4 (Web Text))

Max Activating Token Index: 450

Text #15

Max Range: 5.3085. Min Range: -5.3085

Max Act: 4.8789. Min Act: -0.1700

Data Index: 506370 (C4 (Web Text))

Max Activating Token Index: 914

Text #16

Max Range: 5.3085. Min Range: -5.3085

Max Act: 4.9023. Min Act: -0.1700

Data Index: 142042 (C4 (Web Text))

Max Activating Token Index: 778

Text #17

Max Range: 5.3085. Min Range: -5.3085

Max Act: 4.8476. Min Act: -0.1700

Data Index: 472273 (C4 (Web Text))

Max Activating Token Index: 352

Text #18

Max Range: 5.3085. Min Range: -5.3085

Max Act: 4.8632. Min Act: -0.1700

Data Index: 833688 (C4 (Web Text))

Max Activating Token Index: 368

Text #19

Max Range: 5.3085. Min Range: -5.3085

Max Act: 4.7773. Min Act: -0.1699

Data Index: 763923 (C4 (Web Text))

Max Activating Token Index: 238