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

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

Neuron 256 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: 3.9065. Min Range: -3.9065

Max Act: 3.9065. Min Act: -0.1700

Data Index: 1243514 (C4 (Web Text))

Max Activating Token Index: 534

Text #1

Max Range: 3.9065. Min Range: -3.9065

Max Act: 3.8753. Min Act: -0.1700

Data Index: 741306 (C4 (Web Text))

Max Activating Token Index: 762

Text #2

Max Range: 3.9065. Min Range: -3.9065

Max Act: 3.9065. Min Act: -0.1700

Data Index: 774199 (C4 (Web Text))

Max Activating Token Index: 303

Text #3

Max Range: 3.9065. Min Range: -3.9065

Max Act: 3.7814. Min Act: -0.1700

Data Index: 364388 (C4 (Web Text))

Max Activating Token Index: 278

Text #4

Max Range: 3.9065. Min Range: -3.9065

Max Act: 3.7814. Min Act: -0.1700

Data Index: 207044 (C4 (Web Text))

Max Activating Token Index: 738

Text #5

Max Range: 3.9065. Min Range: -3.9065

Max Act: 3.8127. Min Act: -0.1700

Data Index: 1149970 (C4 (Web Text))

Max Activating Token Index: 848

Text #6

Max Range: 3.9065. Min Range: -3.9065

Max Act: 3.7814. Min Act: -0.1700

Data Index: 735809 (C4 (Web Text))

Max Activating Token Index: 513

Text #7

Max Range: 3.9065. Min Range: -3.9065

Max Act: 3.7501. Min Act: -0.1700

Data Index: 1071261 (C4 (Web Text))

Max Activating Token Index: 446

Text #8

Max Range: 3.9065. Min Range: -3.9065

Max Act: 3.7188. Min Act: -0.1700

Data Index: 820182 (C4 (Web Text))

Max Activating Token Index: 913

Text #9

Max Range: 3.9065. Min Range: -3.9065

Max Act: 3.6876. Min Act: -0.1700

Data Index: 596620 (C4 (Web Text))

Max Activating Token Index: 899

Text #10

Max Range: 3.9065. Min Range: -3.9065

Max Act: 3.6876. Min Act: -0.1700

Data Index: 673323 (C4 (Web Text))

Max Activating Token Index: 823

Text #11

Max Range: 3.9065. Min Range: -3.9065

Max Act: 3.6876. Min Act: -0.1700

Data Index: 147103 (C4 (Web Text))

Max Activating Token Index: 488

Text #12

Max Range: 3.9065. Min Range: -3.9065

Max Act: 3.7188. Min Act: -0.1700

Data Index: 835079 (C4 (Web Text))

Max Activating Token Index: 148

Text #13

Max Range: 3.9065. Min Range: -3.9065

Max Act: 3.7188. Min Act: -0.1700

Data Index: 889472 (C4 (Web Text))

Max Activating Token Index: 592

Text #14

Max Range: 3.9065. Min Range: -3.9065

Max Act: 3.6876. Min Act: -0.1700

Data Index: 419378 (C4 (Web Text))

Max Activating Token Index: 630

Text #15

Max Range: 3.9065. Min Range: -3.9065

Max Act: 3.6876. Min Act: -0.1700

Data Index: 1085816 (C4 (Web Text))

Max Activating Token Index: 956

Text #16

Max Range: 3.9065. Min Range: -3.9065

Max Act: 3.6563. Min Act: -0.1700

Data Index: 1131361 (C4 (Web Text))

Max Activating Token Index: 718

Text #17

Max Range: 3.9065. Min Range: -3.9065

Max Act: 3.6563. Min Act: -0.1700

Data Index: 1221928 (C4 (Web Text))

Max Activating Token Index: 526

Text #18

Max Range: 3.9065. Min Range: -3.9065

Max Act: 3.6249. Min Act: -0.1700

Data Index: 946462 (C4 (Web Text))

Max Activating Token Index: 899

Text #19

Max Range: 3.9065. Min Range: -3.9065

Max Act: 3.6093. Min Act: -0.1700

Data Index: 900556 (C4 (Web Text))

Max Activating Token Index: 902