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

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

Neuron 1922 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: 3.6094. Min Range: -3.6094

Max Act: 3.6094. Min Act: -0.1700

Data Index: 758017 (C4 (Web Text))

Max Activating Token Index: 685

Text #1

Max Range: 3.6094. Min Range: -3.6094

Max Act: 3.5624. Min Act: -0.1700

Data Index: 1051430 (C4 (Web Text))

Max Activating Token Index: 622

Text #2

Max Range: 3.6094. Min Range: -3.6094

Max Act: 3.5468. Min Act: -0.1700

Data Index: 471658 (C4 (Web Text))

Max Activating Token Index: 293

Text #3

Max Range: 3.6094. Min Range: -3.6094

Max Act: 3.5468. Min Act: -0.1700

Data Index: 308092 (C4 (Web Text))

Max Activating Token Index: 832

Text #4

Max Range: 3.6094. Min Range: -3.6094

Max Act: 3.5311. Min Act: -0.1700

Data Index: 622942 (C4 (Web Text))

Max Activating Token Index: 747

Text #5

Max Range: 3.6094. Min Range: -3.6094

Max Act: 3.4684. Min Act: -0.1700

Data Index: 1237791 (C4 (Web Text))

Max Activating Token Index: 859

Text #6

Max Range: 3.6094. Min Range: -3.6094

Max Act: 3.4684. Min Act: -0.1700

Data Index: 22800 (C4 (Web Text))

Max Activating Token Index: 831

Text #7

Max Range: 3.6094. Min Range: -3.6094

Max Act: 3.4371. Min Act: -0.1700

Data Index: 1219796 (C4 (Web Text))

Max Activating Token Index: 22

Text #8

Max Range: 3.6094. Min Range: -3.6094

Max Act: 3.4998. Min Act: -0.1700

Data Index: 587786 (C4 (Web Text))

Max Activating Token Index: 116

Text #9

Max Range: 3.6094. Min Range: -3.6094

Max Act: 3.4371. Min Act: -0.1700

Data Index: 962201 (C4 (Web Text))

Max Activating Token Index: 595

Text #10

Max Range: 3.6094. Min Range: -3.6094

Max Act: 3.4214. Min Act: -0.1700

Data Index: 1339058 (C4 (Web Text))

Max Activating Token Index: 748

Text #11

Max Range: 3.6094. Min Range: -3.6094

Max Act: 3.4214. Min Act: -0.1700

Data Index: 114536 (C4 (Web Text))

Max Activating Token Index: 546

Text #12

Max Range: 3.6094. Min Range: -3.6094

Max Act: 3.4371. Min Act: -0.1700

Data Index: 1275343 (C4 (Web Text))

Max Activating Token Index: 769

Text #13

Max Range: 3.6094. Min Range: -3.6094

Max Act: 3.4057. Min Act: -0.1700

Data Index: 295415 (C4 (Web Text))

Max Activating Token Index: 338

Text #14

Max Range: 3.6094. Min Range: -3.6094

Max Act: 3.4057. Min Act: -0.1700

Data Index: 1248036 (C4 (Web Text))

Max Activating Token Index: 405

Text #15

Max Range: 3.6094. Min Range: -3.6094

Max Act: 3.4371. Min Act: -0.1700

Data Index: 1308792 (C4 (Web Text))

Max Activating Token Index: 72

Text #16

Max Range: 3.6094. Min Range: -3.6094

Max Act: 3.3900. Min Act: -0.1700

Data Index: 624089 (C4 (Web Text))

Max Activating Token Index: 159

Text #17

Max Range: 3.6094. Min Range: -3.6094

Max Act: 3.4057. Min Act: -0.1700

Data Index: 1228291 (C4 (Web Text))

Max Activating Token Index: 788

Text #18

Max Range: 3.6094. Min Range: -3.6094

Max Act: 3.3743. Min Act: -0.1700

Data Index: 1022055 (C4 (Web Text))

Max Activating Token Index: 366

Text #19

Max Range: 3.6094. Min Range: -3.6094

Max Act: 3.3587. Min Act: -0.1700

Data Index: 359181 (C4 (Web Text))

Max Activating Token Index: 287