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

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

Neuron 959 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.8676. Min Range: -3.8676

Max Act: 3.8676. Min Act: -0.1700

Data Index: 913623 (C4 (Web Text))

Max Activating Token Index: 1015

Text #1

Max Range: 3.8676. Min Range: -3.8676

Max Act: 3.7581. Min Act: -0.1700

Data Index: 912218 (C4 (Web Text))

Max Activating Token Index: 1001

Text #2

Max Range: 3.8676. Min Range: -3.8676

Max Act: 3.6172. Min Act: -0.1700

Data Index: 1055714 (C4 (Web Text))

Max Activating Token Index: 994

Text #3

Max Range: 3.8676. Min Range: -3.8676

Max Act: 3.5859. Min Act: -0.1700

Data Index: 965349 (C4 (Web Text))

Max Activating Token Index: 1022

Text #4

Max Range: 3.8676. Min Range: -3.8676

Max Act: 3.5233. Min Act: -0.1700

Data Index: 174040 (C4 (Web Text))

Max Activating Token Index: 973

Text #5

Max Range: 3.8676. Min Range: -3.8676

Max Act: 3.5233. Min Act: -0.1700

Data Index: 798992 (C4 (Web Text))

Max Activating Token Index: 910

Text #6

Max Range: 3.8676. Min Range: -3.8676

Max Act: 3.5233. Min Act: -0.1700

Data Index: 491695 (C4 (Web Text))

Max Activating Token Index: 865

Text #7

Max Range: 3.8676. Min Range: -3.8676

Max Act: 3.4919. Min Act: -0.1700

Data Index: 478991 (C4 (Web Text))

Max Activating Token Index: 1009

Text #8

Max Range: 3.8676. Min Range: -3.8676

Max Act: 3.4919. Min Act: -0.1700

Data Index: 4874 (C4 (Web Text))

Max Activating Token Index: 882

Text #9

Max Range: 3.8676. Min Range: -3.8676

Max Act: 3.4763. Min Act: -0.1700

Data Index: 361915 (C4 (Web Text))

Max Activating Token Index: 962

Text #10

Max Range: 3.8676. Min Range: -3.8676

Max Act: 3.4606. Min Act: -0.1700

Data Index: 1550606 (Python Code)

Max Activating Token Index: 1007

Text #11

Max Range: 3.8676. Min Range: -3.8676

Max Act: 3.4606. Min Act: -0.1700

Data Index: 355493 (C4 (Web Text))

Max Activating Token Index: 882

Text #12

Max Range: 3.8676. Min Range: -3.8676

Max Act: 3.4606. Min Act: -0.1700

Data Index: 796943 (C4 (Web Text))

Max Activating Token Index: 993

Text #13

Max Range: 3.8676. Min Range: -3.8676

Max Act: 3.4292. Min Act: -0.1700

Data Index: 1136961 (C4 (Web Text))

Max Activating Token Index: 1009

Text #14

Max Range: 3.8676. Min Range: -3.8676

Max Act: 3.4292. Min Act: -0.1700

Data Index: 631371 (C4 (Web Text))

Max Activating Token Index: 834

Text #15

Max Range: 3.8676. Min Range: -3.8676

Max Act: 3.4449. Min Act: -0.1700

Data Index: 617150 (C4 (Web Text))

Max Activating Token Index: 888

Text #16

Max Range: 3.8676. Min Range: -3.8676

Max Act: 3.3822. Min Act: -0.1700

Data Index: 237834 (C4 (Web Text))

Max Activating Token Index: 969

Text #17

Max Range: 3.8676. Min Range: -3.8676

Max Act: 3.3979. Min Act: -0.1700

Data Index: 285281 (C4 (Web Text))

Max Activating Token Index: 1000

Text #18

Max Range: 3.8676. Min Range: -3.8676

Max Act: 3.3822. Min Act: -0.1700

Data Index: 1447174 (Python Code)

Max Activating Token Index: 937

Text #19

Max Range: 3.8676. Min Range: -3.8676

Max Act: 3.3351. Min Act: -0.1700

Data Index: 835085 (C4 (Web Text))

Max Activating Token Index: 1001