Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 570 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 3.7403. Min Range: -3.7403

Max Act: 3.7403. Min Act: -0.1700

Data Index: 644049 (C4 (Web Text))

Max Activating Token Index: 714

Text #1

Max Range: 3.7403. Min Range: -3.7403

Max Act: 3.6933. Min Act: -0.1700

Data Index: 1620380 (Python Code)

Max Activating Token Index: 47

Text #2

Max Range: 3.7403. Min Range: -3.7403

Max Act: 3.6933. Min Act: -0.1700

Data Index: 334137 (C4 (Web Text))

Max Activating Token Index: 214

Text #3

Max Range: 3.7403. Min Range: -3.7403

Max Act: 3.7090. Min Act: -0.1700

Data Index: 1432846 (Python Code)

Max Activating Token Index: 467

Text #4

Max Range: 3.7403. Min Range: -3.7403

Max Act: 3.6307. Min Act: -0.1700

Data Index: 245606 (C4 (Web Text))

Max Activating Token Index: 541

Text #5

Max Range: 3.7403. Min Range: -3.7403

Max Act: 3.7246. Min Act: -0.1700

Data Index: 1049726 (C4 (Web Text))

Max Activating Token Index: 551

Text #6

Max Range: 3.7403. Min Range: -3.7403

Max Act: 3.6464. Min Act: -0.1700

Data Index: 1470218 (Python Code)

Max Activating Token Index: 363

Text #7

Max Range: 3.7403. Min Range: -3.7403

Max Act: 3.6777. Min Act: -0.1700

Data Index: 1617571 (Python Code)

Max Activating Token Index: 785

Text #8

Max Range: 3.7403. Min Range: -3.7403

Max Act: 3.6307. Min Act: -0.1700

Data Index: 819576 (C4 (Web Text))

Max Activating Token Index: 272

Text #9

Max Range: 3.7403. Min Range: -3.7403

Max Act: 3.6151. Min Act: -0.1700

Data Index: 1631957 (Python Code)

Max Activating Token Index: 432

Text #10

Max Range: 3.7403. Min Range: -3.7403

Max Act: 3.5838. Min Act: -0.1700

Data Index: 1530835 (Python Code)

Max Activating Token Index: 266

Text #11

Max Range: 3.7403. Min Range: -3.7403

Max Act: 3.6151. Min Act: -0.1700

Data Index: 422527 (C4 (Web Text))

Max Activating Token Index: 608

Text #12

Max Range: 3.7403. Min Range: -3.7403

Max Act: 3.7090. Min Act: -0.1700

Data Index: 138727 (C4 (Web Text))

Max Activating Token Index: 823

Text #13

Max Range: 3.7403. Min Range: -3.7403

Max Act: 3.5524. Min Act: -0.1700

Data Index: 60700 (C4 (Web Text))

Max Activating Token Index: 231

Text #14

Max Range: 3.7403. Min Range: -3.7403

Max Act: 3.5994. Min Act: -0.1700

Data Index: 42290 (C4 (Web Text))

Max Activating Token Index: 410

Text #15

Max Range: 3.7403. Min Range: -3.7403

Max Act: 3.6307. Min Act: -0.1700

Data Index: 1222885 (C4 (Web Text))

Max Activating Token Index: 513

Text #16

Max Range: 3.7403. Min Range: -3.7403

Max Act: 3.5681. Min Act: -0.1700

Data Index: 1574308 (Python Code)

Max Activating Token Index: 663

Text #17

Max Range: 3.7403. Min Range: -3.7403

Max Act: 3.5681. Min Act: -0.1700

Data Index: 1570993 (Python Code)

Max Activating Token Index: 96

Text #18

Max Range: 3.7403. Min Range: -3.7403

Max Act: 3.5681. Min Act: -0.1700

Data Index: 661091 (C4 (Web Text))

Max Activating Token Index: 633

Text #19

Max Range: 3.7403. Min Range: -3.7403

Max Act: 3.5838. Min Act: -0.1700

Data Index: 1591894 (Python Code)

Max Activating Token Index: 408