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

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

Neuron 1113 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: 3.6101. Min Range: -3.6101

Max Act: 3.6101. Min Act: -0.1700

Data Index: 998244 (C4 (Web Text))

Max Activating Token Index: 582

Text #1

Max Range: 3.6101. Min Range: -3.6101

Max Act: 3.5631. Min Act: -0.1700

Data Index: 1180590 (C4 (Web Text))

Max Activating Token Index: 669

Text #2

Max Range: 3.6101. Min Range: -3.6101

Max Act: 3.4848. Min Act: -0.1700

Data Index: 324816 (C4 (Web Text))

Max Activating Token Index: 268

Text #3

Max Range: 3.6101. Min Range: -3.6101

Max Act: 3.3436. Min Act: -0.1700

Data Index: 616864 (C4 (Web Text))

Max Activating Token Index: 482

Text #4

Max Range: 3.6101. Min Range: -3.6101

Max Act: 3.3907. Min Act: -0.1700

Data Index: 1549609 (Python Code)

Max Activating Token Index: 676

Text #5

Max Range: 3.6101. Min Range: -3.6101

Max Act: 3.3907. Min Act: -0.1700

Data Index: 1155428 (C4 (Web Text))

Max Activating Token Index: 883

Text #6

Max Range: 3.6101. Min Range: -3.6101

Max Act: 3.3279. Min Act: -0.1700

Data Index: 1077704 (C4 (Web Text))

Max Activating Token Index: 720

Text #7

Max Range: 3.6101. Min Range: -3.6101

Max Act: 3.3436. Min Act: -0.1700

Data Index: 563085 (C4 (Web Text))

Max Activating Token Index: 460

Text #8

Max Range: 3.6101. Min Range: -3.6101

Max Act: 3.3750. Min Act: -0.1700

Data Index: 833299 (C4 (Web Text))

Max Activating Token Index: 143

Text #9

Max Range: 3.6101. Min Range: -3.6101

Max Act: 3.3436. Min Act: -0.1700

Data Index: 201800 (C4 (Web Text))

Max Activating Token Index: 253

Text #10

Max Range: 3.6101. Min Range: -3.6101

Max Act: 3.2651. Min Act: -0.1700

Data Index: 1370588 (Python Code)

Max Activating Token Index: 593

Text #11

Max Range: 3.6101. Min Range: -3.6101

Max Act: 3.2651. Min Act: -0.1700

Data Index: 1603380 (Python Code)

Max Activating Token Index: 315

Text #12

Max Range: 3.6101. Min Range: -3.6101

Max Act: 3.2494. Min Act: -0.1700

Data Index: 1405557 (Python Code)

Max Activating Token Index: 860

Text #13

Max Range: 3.6101. Min Range: -3.6101

Max Act: 3.2651. Min Act: -0.1700

Data Index: 1310768 (C4 (Web Text))

Max Activating Token Index: 254

Text #14

Max Range: 3.6101. Min Range: -3.6101

Max Act: 3.2337. Min Act: -0.1700

Data Index: 1155175 (C4 (Web Text))

Max Activating Token Index: 355

Text #15

Max Range: 3.6101. Min Range: -3.6101

Max Act: 3.2651. Min Act: -0.1700

Data Index: 1422928 (Python Code)

Max Activating Token Index: 447

Text #16

Max Range: 3.6101. Min Range: -3.6101

Max Act: 3.2494. Min Act: -0.1700

Data Index: 954556 (C4 (Web Text))

Max Activating Token Index: 578

Text #17

Max Range: 3.6101. Min Range: -3.6101

Max Act: 3.2494. Min Act: -0.1700

Data Index: 633387 (C4 (Web Text))

Max Activating Token Index: 709

Text #18

Max Range: 3.6101. Min Range: -3.6101

Max Act: 3.2179. Min Act: -0.1700

Data Index: 1455247 (Python Code)

Max Activating Token Index: 83

Text #19

Max Range: 3.6101. Min Range: -3.6101

Max Act: 3.2337. Min Act: -0.1700

Data Index: 1611485 (Python Code)

Max Activating Token Index: 659