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

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

Neuron 1786 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.5401. Min Range: -3.5401

Max Act: 3.5401. Min Act: -0.1700

Data Index: 118014 (C4 (Web Text))

Max Activating Token Index: 550

Text #1

Max Range: 3.5401. Min Range: -3.5401

Max Act: 3.5088. Min Act: -0.1700

Data Index: 584089 (C4 (Web Text))

Max Activating Token Index: 329

Text #2

Max Range: 3.5401. Min Range: -3.5401

Max Act: 3.5245. Min Act: -0.1700

Data Index: 173620 (C4 (Web Text))

Max Activating Token Index: 42

Text #3

Max Range: 3.5401. Min Range: -3.5401

Max Act: 3.5088. Min Act: -0.1700

Data Index: 926734 (C4 (Web Text))

Max Activating Token Index: 457

Text #4

Max Range: 3.5401. Min Range: -3.5401

Max Act: 3.5088. Min Act: -0.1700

Data Index: 1311336 (C4 (Web Text))

Max Activating Token Index: 386

Text #5

Max Range: 3.5401. Min Range: -3.5401

Max Act: 3.4618. Min Act: -0.1700

Data Index: 1284139 (C4 (Web Text))

Max Activating Token Index: 262

Text #6

Max Range: 3.5401. Min Range: -3.5401

Max Act: 3.4461. Min Act: -0.1700

Data Index: 717012 (C4 (Web Text))

Max Activating Token Index: 710

Text #7

Max Range: 3.5401. Min Range: -3.5401

Max Act: 3.4304. Min Act: -0.1700

Data Index: 367553 (C4 (Web Text))

Max Activating Token Index: 824

Text #8

Max Range: 3.5401. Min Range: -3.5401

Max Act: 3.4461. Min Act: -0.1700

Data Index: 1561739 (Python Code)

Max Activating Token Index: 2

Text #9

Max Range: 3.5401. Min Range: -3.5401

Max Act: 3.4148. Min Act: -0.1700

Data Index: 1181151 (C4 (Web Text))

Max Activating Token Index: 416

Text #10

Max Range: 3.5401. Min Range: -3.5401

Max Act: 3.3834. Min Act: -0.1700

Data Index: 54886 (C4 (Web Text))

Max Activating Token Index: 640

Text #11

Max Range: 3.5401. Min Range: -3.5401

Max Act: 3.3991. Min Act: -0.1700

Data Index: 565739 (C4 (Web Text))

Max Activating Token Index: 489

Text #12

Max Range: 3.5401. Min Range: -3.5401

Max Act: 3.3991. Min Act: -0.1700

Data Index: 535165 (C4 (Web Text))

Max Activating Token Index: 511

Text #13

Max Range: 3.5401. Min Range: -3.5401

Max Act: 3.3520. Min Act: -0.1700

Data Index: 1347753 (C4 (Web Text))

Max Activating Token Index: 1013

Text #14

Max Range: 3.5401. Min Range: -3.5401

Max Act: 3.3520. Min Act: -0.1700

Data Index: 1561327 (Python Code)

Max Activating Token Index: 4

Text #15

Max Range: 3.5401. Min Range: -3.5401

Max Act: 3.3677. Min Act: -0.1700

Data Index: 1421464 (Python Code)

Max Activating Token Index: 2

Text #16

Max Range: 3.5401. Min Range: -3.5401

Max Act: 3.3520. Min Act: -0.1700

Data Index: 1484358 (Python Code)

Max Activating Token Index: 3

Text #17

Max Range: 3.5401. Min Range: -3.5401

Max Act: 3.3520. Min Act: -0.1700

Data Index: 1633605 (Python Code)

Max Activating Token Index: 4

Text #18

Max Range: 3.5401. Min Range: -3.5401

Max Act: 3.3206. Min Act: -0.1700

Data Index: 1606954 (Python Code)

Max Activating Token Index: 5

Text #19

Max Range: 3.5401. Min Range: -3.5401

Max Act: 3.3206. Min Act: -0.1700

Data Index: 1387393 (Python Code)

Max Activating Token Index: 4