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

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

Neuron 1887 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: 5.6010. Min Range: -5.6010

Max Act: 5.6010. Min Act: -0.1700

Data Index: 763396 (C4 (Web Text))

Max Activating Token Index: 451

Text #1

Max Range: 5.6010. Min Range: -5.6010

Max Act: 5.4447. Min Act: -0.1700

Data Index: 194628 (C4 (Web Text))

Max Activating Token Index: 732

Text #2

Max Range: 5.6010. Min Range: -5.6010

Max Act: 5.4135. Min Act: -0.1700

Data Index: 302576 (C4 (Web Text))

Max Activating Token Index: 846

Text #3

Max Range: 5.6010. Min Range: -5.6010

Max Act: 5.2885. Min Act: -0.1700

Data Index: 110178 (C4 (Web Text))

Max Activating Token Index: 77

Text #4

Max Range: 5.6010. Min Range: -5.6010

Max Act: 5.2885. Min Act: -0.1700

Data Index: 710993 (C4 (Web Text))

Max Activating Token Index: 13

Text #5

Max Range: 5.6010. Min Range: -5.6010

Max Act: 5.2885. Min Act: -0.1700

Data Index: 521227 (C4 (Web Text))

Max Activating Token Index: 13

Text #6

Max Range: 5.6010. Min Range: -5.6010

Max Act: 5.1947. Min Act: -0.1700

Data Index: 976964 (C4 (Web Text))

Max Activating Token Index: 729

Text #7

Max Range: 5.6010. Min Range: -5.6010

Max Act: 5.2260. Min Act: -0.1700

Data Index: 1310380 (C4 (Web Text))

Max Activating Token Index: 18

Text #8

Max Range: 5.6010. Min Range: -5.6010

Max Act: 5.2572. Min Act: -0.1700

Data Index: 640198 (C4 (Web Text))

Max Activating Token Index: 1007

Text #9

Max Range: 5.6010. Min Range: -5.6010

Max Act: 5.1635. Min Act: -0.1700

Data Index: 892138 (C4 (Web Text))

Max Activating Token Index: 948

Text #10

Max Range: 5.6010. Min Range: -5.6010

Max Act: 5.1322. Min Act: -0.1700

Data Index: 574304 (C4 (Web Text))

Max Activating Token Index: 930

Text #11

Max Range: 5.6010. Min Range: -5.6010

Max Act: 5.1947. Min Act: -0.1700

Data Index: 353401 (C4 (Web Text))

Max Activating Token Index: 514

Text #12

Max Range: 5.6010. Min Range: -5.6010

Max Act: 5.2260. Min Act: -0.1700

Data Index: 1159531 (C4 (Web Text))

Max Activating Token Index: 335

Text #13

Max Range: 5.6010. Min Range: -5.6010

Max Act: 5.2260. Min Act: -0.1700

Data Index: 817336 (C4 (Web Text))

Max Activating Token Index: 18

Text #14

Max Range: 5.6010. Min Range: -5.6010

Max Act: 5.1010. Min Act: -0.1700

Data Index: 162306 (C4 (Web Text))

Max Activating Token Index: 560

Text #15

Max Range: 5.6010. Min Range: -5.6010

Max Act: 5.1635. Min Act: -0.1700

Data Index: 311532 (C4 (Web Text))

Max Activating Token Index: 28

Text #16

Max Range: 5.6010. Min Range: -5.6010

Max Act: 5.1322. Min Act: -0.1700

Data Index: 1293671 (C4 (Web Text))

Max Activating Token Index: 80

Text #17

Max Range: 5.6010. Min Range: -5.6010

Max Act: 5.1635. Min Act: -0.1700

Data Index: 1061876 (C4 (Web Text))

Max Activating Token Index: 316

Text #18

Max Range: 5.6010. Min Range: -5.6010

Max Act: 5.1322. Min Act: -0.1700

Data Index: 105447 (C4 (Web Text))

Max Activating Token Index: 574

Text #19

Max Range: 5.6010. Min Range: -5.6010

Max Act: 5.1635. Min Act: -0.1700

Data Index: 450749 (C4 (Web Text))

Max Activating Token Index: 473