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

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

Neuron 1492 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.5252. Min Range: -5.5252

Max Act: 5.5252. Min Act: -0.1700

Data Index: 578570 (C4 (Web Text))

Max Activating Token Index: 899

Text #1

Max Range: 5.5252. Min Range: -5.5252

Max Act: 5.4939. Min Act: -0.1700

Data Index: 801150 (C4 (Web Text))

Max Activating Token Index: 574

Text #2

Max Range: 5.5252. Min Range: -5.5252

Max Act: 5.4939. Min Act: -0.1700

Data Index: 929762 (C4 (Web Text))

Max Activating Token Index: 901

Text #3

Max Range: 5.5252. Min Range: -5.5252

Max Act: 5.4627. Min Act: -0.1700

Data Index: 128701 (C4 (Web Text))

Max Activating Token Index: 971

Text #4

Max Range: 5.5252. Min Range: -5.5252

Max Act: 5.4314. Min Act: -0.1700

Data Index: 957583 (C4 (Web Text))

Max Activating Token Index: 326

Text #5

Max Range: 5.5252. Min Range: -5.5252

Max Act: 5.3064. Min Act: -0.1700

Data Index: 769615 (C4 (Web Text))

Max Activating Token Index: 315

Text #6

Max Range: 5.5252. Min Range: -5.5252

Max Act: 5.3689. Min Act: -0.1700

Data Index: 445436 (C4 (Web Text))

Max Activating Token Index: 861

Text #7

Max Range: 5.5252. Min Range: -5.5252

Max Act: 5.3689. Min Act: -0.1700

Data Index: 1204538 (C4 (Web Text))

Max Activating Token Index: 956

Text #8

Max Range: 5.5252. Min Range: -5.5252

Max Act: 5.3064. Min Act: -0.1700

Data Index: 1295642 (C4 (Web Text))

Max Activating Token Index: 1017

Text #9

Max Range: 5.5252. Min Range: -5.5252

Max Act: 5.3064. Min Act: -0.1700

Data Index: 1219536 (C4 (Web Text))

Max Activating Token Index: 581

Text #10

Max Range: 5.5252. Min Range: -5.5252

Max Act: 5.2439. Min Act: -0.1700

Data Index: 1061927 (C4 (Web Text))

Max Activating Token Index: 993

Text #11

Max Range: 5.5252. Min Range: -5.5252

Max Act: 5.2752. Min Act: -0.1700

Data Index: 1028369 (C4 (Web Text))

Max Activating Token Index: 267

Text #12

Max Range: 5.5252. Min Range: -5.5252

Max Act: 5.2752. Min Act: -0.1700

Data Index: 845140 (C4 (Web Text))

Max Activating Token Index: 916

Text #13

Max Range: 5.5252. Min Range: -5.5252

Max Act: 5.1814. Min Act: -0.1700

Data Index: 856339 (C4 (Web Text))

Max Activating Token Index: 232

Text #14

Max Range: 5.5252. Min Range: -5.5252

Max Act: 5.1814. Min Act: -0.1700

Data Index: 1211519 (C4 (Web Text))

Max Activating Token Index: 528

Text #15

Max Range: 5.5252. Min Range: -5.5252

Max Act: 5.2127. Min Act: -0.1700

Data Index: 246800 (C4 (Web Text))

Max Activating Token Index: 824

Text #16

Max Range: 5.5252. Min Range: -5.5252

Max Act: 5.1189. Min Act: -0.1700

Data Index: 916806 (C4 (Web Text))

Max Activating Token Index: 585

Text #17

Max Range: 5.5252. Min Range: -5.5252

Max Act: 5.0877. Min Act: -0.1700

Data Index: 198110 (C4 (Web Text))

Max Activating Token Index: 1010

Text #18

Max Range: 5.5252. Min Range: -5.5252

Max Act: 5.1814. Min Act: -0.1699

Data Index: 697345 (C4 (Web Text))

Max Activating Token Index: 279

Text #19

Max Range: 5.5252. Min Range: -5.5252

Max Act: 5.0564. Min Act: -0.1700

Data Index: 57543 (C4 (Web Text))

Max Activating Token Index: 389