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

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

Neuron 847 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 3.3402. Min Range: -3.3402

Max Act: 3.3402. Min Act: -0.1700

Data Index: 1645881 (Python Code)

Max Activating Token Index: 997

Text #1

Max Range: 3.3402. Min Range: -3.3402

Max Act: 3.2460. Min Act: -0.1700

Data Index: 1054784 (C4 (Web Text))

Max Activating Token Index: 83

Text #2

Max Range: 3.3402. Min Range: -3.3402

Max Act: 3.2145. Min Act: -0.1700

Data Index: 553625 (C4 (Web Text))

Max Activating Token Index: 874

Text #3

Max Range: 3.3402. Min Range: -3.3402

Max Act: 3.1673. Min Act: -0.1700

Data Index: 995723 (C4 (Web Text))

Max Activating Token Index: 961

Text #4

Max Range: 3.3402. Min Range: -3.3402

Max Act: 3.1516. Min Act: -0.1700

Data Index: 244047 (C4 (Web Text))

Max Activating Token Index: 887

Text #5

Max Range: 3.3402. Min Range: -3.3402

Max Act: 3.1358. Min Act: -0.1700

Data Index: 857168 (C4 (Web Text))

Max Activating Token Index: 997

Text #6

Max Range: 3.3402. Min Range: -3.3402

Max Act: 3.0885. Min Act: -0.1700

Data Index: 906768 (C4 (Web Text))

Max Activating Token Index: 904

Text #7

Max Range: 3.3402. Min Range: -3.3402

Max Act: 3.1043. Min Act: -0.1700

Data Index: 1056961 (C4 (Web Text))

Max Activating Token Index: 885

Text #8

Max Range: 3.3402. Min Range: -3.3402

Max Act: 3.1201. Min Act: -0.1700

Data Index: 463443 (C4 (Web Text))

Max Activating Token Index: 932

Text #9

Max Range: 3.3402. Min Range: -3.3402

Max Act: 3.1043. Min Act: -0.1700

Data Index: 1259544 (C4 (Web Text))

Max Activating Token Index: 36

Text #10

Max Range: 3.3402. Min Range: -3.3402

Max Act: 3.0885. Min Act: -0.1700

Data Index: 498812 (C4 (Web Text))

Max Activating Token Index: 950

Text #11

Max Range: 3.3402. Min Range: -3.3402

Max Act: 3.1043. Min Act: -0.1700

Data Index: 852478 (C4 (Web Text))

Max Activating Token Index: 774

Text #12

Max Range: 3.3402. Min Range: -3.3402

Max Act: 3.1043. Min Act: -0.1700

Data Index: 426029 (C4 (Web Text))

Max Activating Token Index: 395

Text #13

Max Range: 3.3402. Min Range: -3.3402

Max Act: 3.1043. Min Act: -0.1700

Data Index: 1473 (C4 (Web Text))

Max Activating Token Index: 986

Text #14

Max Range: 3.3402. Min Range: -3.3402

Max Act: 3.0728. Min Act: -0.1700

Data Index: 200357 (C4 (Web Text))

Max Activating Token Index: 841

Text #15

Max Range: 3.3402. Min Range: -3.3402

Max Act: 3.0728. Min Act: -0.1700

Data Index: 1450930 (Python Code)

Max Activating Token Index: 671

Text #16

Max Range: 3.3402. Min Range: -3.3402

Max Act: 3.0728. Min Act: -0.1700

Data Index: 1335486 (C4 (Web Text))

Max Activating Token Index: 359

Text #17

Max Range: 3.3402. Min Range: -3.3402

Max Act: 3.0570. Min Act: -0.1700

Data Index: 482619 (C4 (Web Text))

Max Activating Token Index: 193

Text #18

Max Range: 3.3402. Min Range: -3.3402

Max Act: 3.0412. Min Act: -0.1700

Data Index: 752441 (C4 (Web Text))

Max Activating Token Index: 242

Text #19

Max Range: 3.3402. Min Range: -3.3402

Max Act: 3.0412. Min Act: -0.1700

Data Index: 883335 (C4 (Web Text))

Max Activating Token Index: 143