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

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

Neuron 1534 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.3079. Min Range: -3.3079

Max Act: 3.3079. Min Act: -0.1700

Data Index: 1205617 (C4 (Web Text))

Max Activating Token Index: 308

Text #1

Max Range: 3.3079. Min Range: -3.3079

Max Act: 3.2921. Min Act: -0.1700

Data Index: 42310 (C4 (Web Text))

Max Activating Token Index: 197

Text #2

Max Range: 3.3079. Min Range: -3.3079

Max Act: 3.2450. Min Act: -0.1700

Data Index: 1279985 (C4 (Web Text))

Max Activating Token Index: 84

Text #3

Max Range: 3.3079. Min Range: -3.3079

Max Act: 3.2450. Min Act: -0.1700

Data Index: 246537 (C4 (Web Text))

Max Activating Token Index: 61

Text #4

Max Range: 3.3079. Min Range: -3.3079

Max Act: 3.2293. Min Act: -0.1700

Data Index: 265460 (C4 (Web Text))

Max Activating Token Index: 75

Text #5

Max Range: 3.3079. Min Range: -3.3079

Max Act: 3.2293. Min Act: -0.1700

Data Index: 1230177 (C4 (Web Text))

Max Activating Token Index: 43

Text #6

Max Range: 3.3079. Min Range: -3.3079

Max Act: 3.2136. Min Act: -0.1700

Data Index: 494477 (C4 (Web Text))

Max Activating Token Index: 161

Text #7

Max Range: 3.3079. Min Range: -3.3079

Max Act: 3.2136. Min Act: -0.1700

Data Index: 921152 (C4 (Web Text))

Max Activating Token Index: 606

Text #8

Max Range: 3.3079. Min Range: -3.3079

Max Act: 3.2136. Min Act: -0.1700

Data Index: 108051 (C4 (Web Text))

Max Activating Token Index: 495

Text #9

Max Range: 3.3079. Min Range: -3.3079

Max Act: 3.1978. Min Act: -0.1700

Data Index: 692816 (C4 (Web Text))

Max Activating Token Index: 698

Text #10

Max Range: 3.3079. Min Range: -3.3079

Max Act: 3.1978. Min Act: -0.1700

Data Index: 420981 (C4 (Web Text))

Max Activating Token Index: 189

Text #11

Max Range: 3.3079. Min Range: -3.3079

Max Act: 3.1821. Min Act: -0.1700

Data Index: 247302 (C4 (Web Text))

Max Activating Token Index: 187

Text #12

Max Range: 3.3079. Min Range: -3.3079

Max Act: 3.1821. Min Act: -0.1700

Data Index: 237441 (C4 (Web Text))

Max Activating Token Index: 771

Text #13

Max Range: 3.3079. Min Range: -3.3079

Max Act: 3.1821. Min Act: -0.1700

Data Index: 31879 (C4 (Web Text))

Max Activating Token Index: 18

Text #14

Max Range: 3.3079. Min Range: -3.3079

Max Act: 3.1663. Min Act: -0.1700

Data Index: 331405 (C4 (Web Text))

Max Activating Token Index: 618

Text #15

Max Range: 3.3079. Min Range: -3.3079

Max Act: 3.1663. Min Act: -0.1700

Data Index: 682781 (C4 (Web Text))

Max Activating Token Index: 232

Text #16

Max Range: 3.3079. Min Range: -3.3079

Max Act: 3.1663. Min Act: -0.1700

Data Index: 1209116 (C4 (Web Text))

Max Activating Token Index: 841

Text #17

Max Range: 3.3079. Min Range: -3.3079

Max Act: 3.1821. Min Act: -0.1700

Data Index: 200232 (C4 (Web Text))

Max Activating Token Index: 475

Text #18

Max Range: 3.3079. Min Range: -3.3079

Max Act: 3.1821. Min Act: -0.1700

Data Index: 167623 (C4 (Web Text))

Max Activating Token Index: 85

Text #19

Max Range: 3.3079. Min Range: -3.3079

Max Act: 3.1348. Min Act: -0.1700

Data Index: 907443 (C4 (Web Text))

Max Activating Token Index: 793