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

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

Neuron 1808 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.9947. Min Range: -3.9947

Max Act: 3.9947. Min Act: -0.1700

Data Index: 1037834 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 3.9947. Min Range: -3.9947

Max Act: 3.9322. Min Act: -0.1700

Data Index: 520690 (C4 (Web Text))

Max Activating Token Index: 919

Text #2

Max Range: 3.9947. Min Range: -3.9947

Max Act: 3.8384. Min Act: -0.1700

Data Index: 1133698 (C4 (Web Text))

Max Activating Token Index: 889

Text #3

Max Range: 3.9947. Min Range: -3.9947

Max Act: 3.8071. Min Act: -0.1700

Data Index: 653842 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 3.9947. Min Range: -3.9947

Max Act: 3.6819. Min Act: -0.1700

Data Index: 1493232 (Python Code)

Max Activating Token Index: 1023

Text #5

Max Range: 3.9947. Min Range: -3.9947

Max Act: 3.5880. Min Act: -0.1700

Data Index: 1358981 (C4 (Web Text))

Max Activating Token Index: 262

Text #6

Max Range: 3.9947. Min Range: -3.9947

Max Act: 3.5880. Min Act: -0.1700

Data Index: 1441933 (Python Code)

Max Activating Token Index: 1023

Text #7

Max Range: 3.9947. Min Range: -3.9947

Max Act: 3.5880. Min Act: -0.1700

Data Index: 1325966 (C4 (Web Text))

Max Activating Token Index: 390

Text #8

Max Range: 3.9947. Min Range: -3.9947

Max Act: 3.4940. Min Act: -0.1700

Data Index: 114998 (C4 (Web Text))

Max Activating Token Index: 820

Text #9

Max Range: 3.9947. Min Range: -3.9947

Max Act: 3.5253. Min Act: -0.1700

Data Index: 1388631 (Python Code)

Max Activating Token Index: 1023

Text #10

Max Range: 3.9947. Min Range: -3.9947

Max Act: 3.5567. Min Act: -0.1700

Data Index: 992124 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 3.9947. Min Range: -3.9947

Max Act: 3.4940. Min Act: -0.1700

Data Index: 817285 (C4 (Web Text))

Max Activating Token Index: 75

Text #12

Max Range: 3.9947. Min Range: -3.9947

Max Act: 3.5567. Min Act: -0.1700

Data Index: 151236 (C4 (Web Text))

Max Activating Token Index: 869

Text #13

Max Range: 3.9947. Min Range: -3.9947

Max Act: 3.5253. Min Act: -0.1700

Data Index: 794006 (C4 (Web Text))

Max Activating Token Index: 55

Text #14

Max Range: 3.9947. Min Range: -3.9947

Max Act: 3.4940. Min Act: -0.1700

Data Index: 499149 (C4 (Web Text))

Max Activating Token Index: 1023

Text #15

Max Range: 3.9947. Min Range: -3.9947

Max Act: 3.5253. Min Act: -0.1700

Data Index: 383269 (C4 (Web Text))

Max Activating Token Index: 1023

Text #16

Max Range: 3.9947. Min Range: -3.9947

Max Act: 3.4627. Min Act: -0.1700

Data Index: 374994 (C4 (Web Text))

Max Activating Token Index: 557

Text #17

Max Range: 3.9947. Min Range: -3.9947

Max Act: 3.4940. Min Act: -0.1700

Data Index: 1135779 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 3.9947. Min Range: -3.9947

Max Act: 3.4940. Min Act: -0.1700

Data Index: 350432 (C4 (Web Text))

Max Activating Token Index: 229

Text #19

Max Range: 3.9947. Min Range: -3.9947

Max Act: 3.4940. Min Act: -0.1700

Data Index: 134004 (C4 (Web Text))

Max Activating Token Index: 784