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

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

Neuron 350 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.2897. Min Range: -5.2897

Max Act: 5.2897. Min Act: -0.1700

Data Index: 497050 (C4 (Web Text))

Max Activating Token Index: 665

Text #1

Max Range: 5.2897. Min Range: -5.2897

Max Act: 5.0710. Min Act: -0.1700

Data Index: 88112 (C4 (Web Text))

Max Activating Token Index: 464

Text #2

Max Range: 5.2897. Min Range: -5.2897

Max Act: 5.0085. Min Act: -0.1700

Data Index: 1258824 (C4 (Web Text))

Max Activating Token Index: 670

Text #3

Max Range: 5.2897. Min Range: -5.2897

Max Act: 5.0397. Min Act: -0.1700

Data Index: 635299 (C4 (Web Text))

Max Activating Token Index: 460

Text #4

Max Range: 5.2897. Min Range: -5.2897

Max Act: 5.0710. Min Act: -0.1700

Data Index: 93636 (C4 (Web Text))

Max Activating Token Index: 91

Text #5

Max Range: 5.2897. Min Range: -5.2897

Max Act: 5.0397. Min Act: -0.1700

Data Index: 558457 (C4 (Web Text))

Max Activating Token Index: 408

Text #6

Max Range: 5.2897. Min Range: -5.2897

Max Act: 5.0397. Min Act: -0.1700

Data Index: 153163 (C4 (Web Text))

Max Activating Token Index: 407

Text #7

Max Range: 5.2897. Min Range: -5.2897

Max Act: 5.0085. Min Act: -0.1700

Data Index: 1237551 (C4 (Web Text))

Max Activating Token Index: 898

Text #8

Max Range: 5.2897. Min Range: -5.2897

Max Act: 4.9460. Min Act: -0.1700

Data Index: 444686 (C4 (Web Text))

Max Activating Token Index: 433

Text #9

Max Range: 5.2897. Min Range: -5.2897

Max Act: 5.0085. Min Act: -0.1700

Data Index: 412751 (C4 (Web Text))

Max Activating Token Index: 463

Text #10

Max Range: 5.2897. Min Range: -5.2897

Max Act: 4.9772. Min Act: -0.1700

Data Index: 48034 (C4 (Web Text))

Max Activating Token Index: 416

Text #11

Max Range: 5.2897. Min Range: -5.2897

Max Act: 4.9147. Min Act: -0.1700

Data Index: 479078 (C4 (Web Text))

Max Activating Token Index: 453

Text #12

Max Range: 5.2897. Min Range: -5.2897

Max Act: 4.8835. Min Act: -0.1700

Data Index: 5050 (C4 (Web Text))

Max Activating Token Index: 238

Text #13

Max Range: 5.2897. Min Range: -5.2897

Max Act: 4.9460. Min Act: -0.1700

Data Index: 135488 (C4 (Web Text))

Max Activating Token Index: 533

Text #14

Max Range: 5.2897. Min Range: -5.2897

Max Act: 4.9460. Min Act: -0.1700

Data Index: 20996 (C4 (Web Text))

Max Activating Token Index: 640

Text #15

Max Range: 5.2897. Min Range: -5.2897

Max Act: 4.9147. Min Act: -0.1700

Data Index: 69804 (C4 (Web Text))

Max Activating Token Index: 402

Text #16

Max Range: 5.2897. Min Range: -5.2897

Max Act: 4.9147. Min Act: -0.1700

Data Index: 70346 (C4 (Web Text))

Max Activating Token Index: 447

Text #17

Max Range: 5.2897. Min Range: -5.2897

Max Act: 4.9460. Min Act: -0.1700

Data Index: 774805 (C4 (Web Text))

Max Activating Token Index: 788

Text #18

Max Range: 5.2897. Min Range: -5.2897

Max Act: 4.9147. Min Act: -0.1700

Data Index: 270376 (C4 (Web Text))

Max Activating Token Index: 90

Text #19

Max Range: 5.2897. Min Range: -5.2897

Max Act: 4.9460. Min Act: -0.1700

Data Index: 588834 (C4 (Web Text))

Max Activating Token Index: 682