Model: GELU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1656 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 4.2191. Min Range: -4.2191

Max Act: 4.2191. Min Act: -0.1700

Data Index: 231608 (C4 (Web Text))

Max Activating Token Index: 970

Text #1

Max Range: 4.2191. Min Range: -4.2191

Max Act: 4.0941. Min Act: -0.1700

Data Index: 1032310 (C4 (Web Text))

Max Activating Token Index: 55

Text #2

Max Range: 4.2191. Min Range: -4.2191

Max Act: 3.9690. Min Act: -0.1700

Data Index: 307161 (C4 (Web Text))

Max Activating Token Index: 641

Text #3

Max Range: 4.2191. Min Range: -4.2191

Max Act: 3.9377. Min Act: -0.1700

Data Index: 589830 (C4 (Web Text))

Max Activating Token Index: 463

Text #4

Max Range: 4.2191. Min Range: -4.2191

Max Act: 3.9065. Min Act: -0.1700

Data Index: 1119746 (C4 (Web Text))

Max Activating Token Index: 390

Text #5

Max Range: 4.2191. Min Range: -4.2191

Max Act: 3.8439. Min Act: -0.1700

Data Index: 1302833 (C4 (Web Text))

Max Activating Token Index: 112

Text #6

Max Range: 4.2191. Min Range: -4.2191

Max Act: 3.8439. Min Act: -0.1700

Data Index: 745274 (C4 (Web Text))

Max Activating Token Index: 647

Text #7

Max Range: 4.2191. Min Range: -4.2191

Max Act: 3.8439. Min Act: -0.1700

Data Index: 1126936 (C4 (Web Text))

Max Activating Token Index: 685

Text #8

Max Range: 4.2191. Min Range: -4.2191

Max Act: 3.8439. Min Act: -0.1700

Data Index: 501639 (C4 (Web Text))

Max Activating Token Index: 920

Text #9

Max Range: 4.2191. Min Range: -4.2191

Max Act: 3.8126. Min Act: -0.1700

Data Index: 839935 (C4 (Web Text))

Max Activating Token Index: 971

Text #10

Max Range: 4.2191. Min Range: -4.2191

Max Act: 3.8126. Min Act: -0.1700

Data Index: 1309802 (C4 (Web Text))

Max Activating Token Index: 592

Text #11

Max Range: 4.2191. Min Range: -4.2191

Max Act: 3.7813. Min Act: -0.1700

Data Index: 906946 (C4 (Web Text))

Max Activating Token Index: 355

Text #12

Max Range: 4.2191. Min Range: -4.2191

Max Act: 3.7188. Min Act: -0.1700

Data Index: 1617144 (Python Code)

Max Activating Token Index: 744

Text #13

Max Range: 4.2191. Min Range: -4.2191

Max Act: 3.6875. Min Act: -0.1700

Data Index: 988100 (C4 (Web Text))

Max Activating Token Index: 152

Text #14

Max Range: 4.2191. Min Range: -4.2191

Max Act: 3.6718. Min Act: -0.1700

Data Index: 310928 (C4 (Web Text))

Max Activating Token Index: 91

Text #15

Max Range: 4.2191. Min Range: -4.2191

Max Act: 3.6718. Min Act: -0.1700

Data Index: 223743 (C4 (Web Text))

Max Activating Token Index: 735

Text #16

Max Range: 4.2191. Min Range: -4.2191

Max Act: 3.6875. Min Act: -0.1700

Data Index: 1195465 (C4 (Web Text))

Max Activating Token Index: 835

Text #17

Max Range: 4.2191. Min Range: -4.2191

Max Act: 3.6249. Min Act: -0.1700

Data Index: 1039653 (C4 (Web Text))

Max Activating Token Index: 414

Text #18

Max Range: 4.2191. Min Range: -4.2191

Max Act: 3.6405. Min Act: -0.1700

Data Index: 742552 (C4 (Web Text))

Max Activating Token Index: 297

Text #19

Max Range: 4.2191. Min Range: -4.2191

Max Act: 3.6092. Min Act: -0.1700

Data Index: 316168 (C4 (Web Text))

Max Activating Token Index: 54