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

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

Neuron 403 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: 3.8545. Min Range: -3.8545

Max Act: 3.8545. Min Act: -0.1700

Data Index: 116153 (C4 (Web Text))

Max Activating Token Index: 683

Text #1

Max Range: 3.8545. Min Range: -3.8545

Max Act: 3.8389. Min Act: -0.1700

Data Index: 1082409 (C4 (Web Text))

Max Activating Token Index: 571

Text #2

Max Range: 3.8545. Min Range: -3.8545

Max Act: 3.7763. Min Act: -0.1700

Data Index: 566493 (C4 (Web Text))

Max Activating Token Index: 731

Text #3

Max Range: 3.8545. Min Range: -3.8545

Max Act: 3.8389. Min Act: -0.1700

Data Index: 1208949 (C4 (Web Text))

Max Activating Token Index: 820

Text #4

Max Range: 3.8545. Min Range: -3.8545

Max Act: 3.7607. Min Act: -0.1700

Data Index: 1194211 (C4 (Web Text))

Max Activating Token Index: 532

Text #5

Max Range: 3.8545. Min Range: -3.8545

Max Act: 3.8545. Min Act: -0.1700

Data Index: 472538 (C4 (Web Text))

Max Activating Token Index: 582

Text #6

Max Range: 3.8545. Min Range: -3.8545

Max Act: 3.8076. Min Act: -0.1700

Data Index: 821825 (C4 (Web Text))

Max Activating Token Index: 894

Text #7

Max Range: 3.8545. Min Range: -3.8545

Max Act: 3.8232. Min Act: -0.1700

Data Index: 1063102 (C4 (Web Text))

Max Activating Token Index: 562

Text #8

Max Range: 3.8545. Min Range: -3.8545

Max Act: 3.8076. Min Act: -0.1700

Data Index: 394629 (C4 (Web Text))

Max Activating Token Index: 805

Text #9

Max Range: 3.8545. Min Range: -3.8545

Max Act: 3.7763. Min Act: -0.1700

Data Index: 1096506 (C4 (Web Text))

Max Activating Token Index: 546

Text #10

Max Range: 3.8545. Min Range: -3.8545

Max Act: 3.7607. Min Act: -0.1700

Data Index: 566451 (C4 (Web Text))

Max Activating Token Index: 255

Text #11

Max Range: 3.8545. Min Range: -3.8545

Max Act: 3.7294. Min Act: -0.1700

Data Index: 83405 (C4 (Web Text))

Max Activating Token Index: 551

Text #12

Max Range: 3.8545. Min Range: -3.8545

Max Act: 3.7607. Min Act: -0.1700

Data Index: 245300 (C4 (Web Text))

Max Activating Token Index: 484

Text #13

Max Range: 3.8545. Min Range: -3.8545

Max Act: 3.7294. Min Act: -0.1700

Data Index: 503 (C4 (Web Text))

Max Activating Token Index: 524

Text #14

Max Range: 3.8545. Min Range: -3.8545

Max Act: 3.7294. Min Act: -0.1700

Data Index: 248324 (C4 (Web Text))

Max Activating Token Index: 814

Text #15

Max Range: 3.8545. Min Range: -3.8545

Max Act: 3.7137. Min Act: -0.1700

Data Index: 250036 (C4 (Web Text))

Max Activating Token Index: 476

Text #16

Max Range: 3.8545. Min Range: -3.8545

Max Act: 3.6981. Min Act: -0.1700

Data Index: 264134 (C4 (Web Text))

Max Activating Token Index: 403

Text #17

Max Range: 3.8545. Min Range: -3.8545

Max Act: 3.7137. Min Act: -0.1700

Data Index: 604793 (C4 (Web Text))

Max Activating Token Index: 581

Text #18

Max Range: 3.8545. Min Range: -3.8545

Max Act: 3.7450. Min Act: -0.1700

Data Index: 281812 (C4 (Web Text))

Max Activating Token Index: 985

Text #19

Max Range: 3.8545. Min Range: -3.8545

Max Act: 3.7294. Min Act: -0.1700

Data Index: 513702 (C4 (Web Text))

Max Activating Token Index: 542