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

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

Neuron 512 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: 3.4853. Min Range: -3.4853

Max Act: 3.4853. Min Act: -0.1700

Data Index: 306760 (C4 (Web Text))

Max Activating Token Index: 863

Text #1

Max Range: 3.4853. Min Range: -3.4853

Max Act: 3.4853. Min Act: -0.1700

Data Index: 977780 (C4 (Web Text))

Max Activating Token Index: 928

Text #2

Max Range: 3.4853. Min Range: -3.4853

Max Act: 3.4696. Min Act: -0.1700

Data Index: 657633 (C4 (Web Text))

Max Activating Token Index: 532

Text #3

Max Range: 3.4853. Min Range: -3.4853

Max Act: 3.4382. Min Act: -0.1700

Data Index: 615 (C4 (Web Text))

Max Activating Token Index: 389

Text #4

Max Range: 3.4853. Min Range: -3.4853

Max Act: 3.4539. Min Act: -0.1700

Data Index: 480972 (C4 (Web Text))

Max Activating Token Index: 839

Text #5

Max Range: 3.4853. Min Range: -3.4853

Max Act: 3.4382. Min Act: -0.1700

Data Index: 18721 (C4 (Web Text))

Max Activating Token Index: 231

Text #6

Max Range: 3.4853. Min Range: -3.4853

Max Act: 3.4382. Min Act: -0.1700

Data Index: 1328368 (C4 (Web Text))

Max Activating Token Index: 565

Text #7

Max Range: 3.4853. Min Range: -3.4853

Max Act: 3.4226. Min Act: -0.1700

Data Index: 1042251 (C4 (Web Text))

Max Activating Token Index: 977

Text #8

Max Range: 3.4853. Min Range: -3.4853

Max Act: 3.4069. Min Act: -0.1700

Data Index: 983933 (C4 (Web Text))

Max Activating Token Index: 9

Text #9

Max Range: 3.4853. Min Range: -3.4853

Max Act: 3.4226. Min Act: -0.1700

Data Index: 373313 (C4 (Web Text))

Max Activating Token Index: 711

Text #10

Max Range: 3.4853. Min Range: -3.4853

Max Act: 3.4069. Min Act: -0.1700

Data Index: 147213 (C4 (Web Text))

Max Activating Token Index: 865

Text #11

Max Range: 3.4853. Min Range: -3.4853

Max Act: 3.3598. Min Act: -0.1700

Data Index: 1229860 (C4 (Web Text))

Max Activating Token Index: 11

Text #12

Max Range: 3.4853. Min Range: -3.4853

Max Act: 3.3598. Min Act: -0.1700

Data Index: 1306144 (C4 (Web Text))

Max Activating Token Index: 33

Text #13

Max Range: 3.4853. Min Range: -3.4853

Max Act: 3.3441. Min Act: -0.1700

Data Index: 689225 (C4 (Web Text))

Max Activating Token Index: 858

Text #14

Max Range: 3.4853. Min Range: -3.4853

Max Act: 3.3441. Min Act: -0.1700

Data Index: 620354 (C4 (Web Text))

Max Activating Token Index: 139

Text #15

Max Range: 3.4853. Min Range: -3.4853

Max Act: 3.3441. Min Act: -0.1700

Data Index: 1074664 (C4 (Web Text))

Max Activating Token Index: 643

Text #16

Max Range: 3.4853. Min Range: -3.4853

Max Act: 3.3284. Min Act: -0.1700

Data Index: 433081 (C4 (Web Text))

Max Activating Token Index: 326

Text #17

Max Range: 3.4853. Min Range: -3.4853

Max Act: 3.3284. Min Act: -0.1700

Data Index: 1187811 (C4 (Web Text))

Max Activating Token Index: 537

Text #18

Max Range: 3.4853. Min Range: -3.4853

Max Act: 3.2970. Min Act: -0.1700

Data Index: 1017822 (C4 (Web Text))

Max Activating Token Index: 404

Text #19

Max Range: 3.4853. Min Range: -3.4853

Max Act: 3.3284. Min Act: -0.1700

Data Index: 79547 (C4 (Web Text))

Max Activating Token Index: 631