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

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

Neuron 1858 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: 5.8573. Min Range: -5.8573

Max Act: 5.8573. Min Act: -0.1700

Data Index: 974671 (C4 (Web Text))

Max Activating Token Index: 833

Text #1

Max Range: 5.8573. Min Range: -5.8573

Max Act: 5.7323. Min Act: -0.1700

Data Index: 102647 (C4 (Web Text))

Max Activating Token Index: 11

Text #2

Max Range: 5.8573. Min Range: -5.8573

Max Act: 5.7323. Min Act: -0.1700

Data Index: 1060380 (C4 (Web Text))

Max Activating Token Index: 790

Text #3

Max Range: 5.8573. Min Range: -5.8573

Max Act: 5.6386. Min Act: -0.1699

Data Index: 987696 (C4 (Web Text))

Max Activating Token Index: 242

Text #4

Max Range: 5.8573. Min Range: -5.8573

Max Act: 5.6386. Min Act: -0.1700

Data Index: 268895 (C4 (Web Text))

Max Activating Token Index: 943

Text #5

Max Range: 5.8573. Min Range: -5.8573

Max Act: 5.5761. Min Act: -0.1700

Data Index: 1273248 (C4 (Web Text))

Max Activating Token Index: 953

Text #6

Max Range: 5.8573. Min Range: -5.8573

Max Act: 5.6386. Min Act: -0.1700

Data Index: 1305166 (C4 (Web Text))

Max Activating Token Index: 976

Text #7

Max Range: 5.8573. Min Range: -5.8573

Max Act: 5.5761. Min Act: -0.1700

Data Index: 414034 (C4 (Web Text))

Max Activating Token Index: 1004

Text #8

Max Range: 5.8573. Min Range: -5.8573

Max Act: 5.5761. Min Act: -0.1700

Data Index: 158257 (C4 (Web Text))

Max Activating Token Index: 603

Text #9

Max Range: 5.8573. Min Range: -5.8573

Max Act: 5.5448. Min Act: -0.1700

Data Index: 786585 (C4 (Web Text))

Max Activating Token Index: 356

Text #10

Max Range: 5.8573. Min Range: -5.8573

Max Act: 5.5136. Min Act: -0.1700

Data Index: 175205 (C4 (Web Text))

Max Activating Token Index: 365

Text #11

Max Range: 5.8573. Min Range: -5.8573

Max Act: 5.4823. Min Act: -0.1700

Data Index: 705942 (C4 (Web Text))

Max Activating Token Index: 566

Text #12

Max Range: 5.8573. Min Range: -5.8573

Max Act: 5.4823. Min Act: -0.1700

Data Index: 388829 (C4 (Web Text))

Max Activating Token Index: 696

Text #13

Max Range: 5.8573. Min Range: -5.8573

Max Act: 5.5136. Min Act: -0.1700

Data Index: 1071033 (C4 (Web Text))

Max Activating Token Index: 544

Text #14

Max Range: 5.8573. Min Range: -5.8573

Max Act: 5.4823. Min Act: -0.1700

Data Index: 828508 (C4 (Web Text))

Max Activating Token Index: 894

Text #15

Max Range: 5.8573. Min Range: -5.8573

Max Act: 5.4823. Min Act: -0.1700

Data Index: 1096004 (C4 (Web Text))

Max Activating Token Index: 562

Text #16

Max Range: 5.8573. Min Range: -5.8573

Max Act: 5.4823. Min Act: -0.1699

Data Index: 338731 (C4 (Web Text))

Max Activating Token Index: 970

Text #17

Max Range: 5.8573. Min Range: -5.8573

Max Act: 5.4511. Min Act: -0.1700

Data Index: 1180070 (C4 (Web Text))

Max Activating Token Index: 603

Text #18

Max Range: 5.8573. Min Range: -5.8573

Max Act: 5.4511. Min Act: -0.1700

Data Index: 551881 (C4 (Web Text))

Max Activating Token Index: 661

Text #19

Max Range: 5.8573. Min Range: -5.8573

Max Act: 5.4823. Min Act: -0.1700

Data Index: 569328 (C4 (Web Text))

Max Activating Token Index: 936