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

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

Neuron 1453 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.6046. Min Range: -3.6046

Max Act: 3.6046. Min Act: -0.1700

Data Index: 50303 (C4 (Web Text))

Max Activating Token Index: 498

Text #1

Max Range: 3.6046. Min Range: -3.6046

Max Act: 3.5576. Min Act: -0.1700

Data Index: 600042 (C4 (Web Text))

Max Activating Token Index: 115

Text #2

Max Range: 3.6046. Min Range: -3.6046

Max Act: 3.4950. Min Act: -0.1700

Data Index: 167281 (C4 (Web Text))

Max Activating Token Index: 607

Text #3

Max Range: 3.6046. Min Range: -3.6046

Max Act: 3.5106. Min Act: -0.1700

Data Index: 570962 (C4 (Web Text))

Max Activating Token Index: 989

Text #4

Max Range: 3.6046. Min Range: -3.6046

Max Act: 3.5106. Min Act: -0.1700

Data Index: 1197185 (C4 (Web Text))

Max Activating Token Index: 657

Text #5

Max Range: 3.6046. Min Range: -3.6046

Max Act: 3.5106. Min Act: -0.1700

Data Index: 1204957 (C4 (Web Text))

Max Activating Token Index: 382

Text #6

Max Range: 3.6046. Min Range: -3.6046

Max Act: 3.4636. Min Act: -0.1700

Data Index: 885780 (C4 (Web Text))

Max Activating Token Index: 142

Text #7

Max Range: 3.6046. Min Range: -3.6046

Max Act: 3.4793. Min Act: -0.1700

Data Index: 717516 (C4 (Web Text))

Max Activating Token Index: 367

Text #8

Max Range: 3.6046. Min Range: -3.6046

Max Act: 3.4950. Min Act: -0.1700

Data Index: 851261 (C4 (Web Text))

Max Activating Token Index: 119

Text #9

Max Range: 3.6046. Min Range: -3.6046

Max Act: 3.4793. Min Act: -0.1700

Data Index: 859278 (C4 (Web Text))

Max Activating Token Index: 15

Text #10

Max Range: 3.6046. Min Range: -3.6046

Max Act: 3.4480. Min Act: -0.1700

Data Index: 1207883 (C4 (Web Text))

Max Activating Token Index: 62

Text #11

Max Range: 3.6046. Min Range: -3.6046

Max Act: 3.4323. Min Act: -0.1700

Data Index: 1025876 (C4 (Web Text))

Max Activating Token Index: 732

Text #12

Max Range: 3.6046. Min Range: -3.6046

Max Act: 3.4323. Min Act: -0.1700

Data Index: 377653 (C4 (Web Text))

Max Activating Token Index: 887

Text #13

Max Range: 3.6046. Min Range: -3.6046

Max Act: 3.4166. Min Act: -0.1700

Data Index: 604945 (C4 (Web Text))

Max Activating Token Index: 103

Text #14

Max Range: 3.6046. Min Range: -3.6046

Max Act: 3.4166. Min Act: -0.1700

Data Index: 1076975 (C4 (Web Text))

Max Activating Token Index: 579

Text #15

Max Range: 3.6046. Min Range: -3.6046

Max Act: 3.4323. Min Act: -0.1700

Data Index: 883906 (C4 (Web Text))

Max Activating Token Index: 33

Text #16

Max Range: 3.6046. Min Range: -3.6046

Max Act: 3.4009. Min Act: -0.1700

Data Index: 152336 (C4 (Web Text))

Max Activating Token Index: 764

Text #17

Max Range: 3.6046. Min Range: -3.6046

Max Act: 3.4323. Min Act: -0.1700

Data Index: 1250121 (C4 (Web Text))

Max Activating Token Index: 803

Text #18

Max Range: 3.6046. Min Range: -3.6046

Max Act: 3.4480. Min Act: -0.1700

Data Index: 1372388 (Python Code)

Max Activating Token Index: 1021

Text #19

Max Range: 3.6046. Min Range: -3.6046

Max Act: 3.4009. Min Act: -0.1700

Data Index: 272915 (C4 (Web Text))

Max Activating Token Index: 810