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

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

Neuron 1938 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: 3.3938. Min Range: -3.3938

Max Act: 3.3938. Min Act: -0.1700

Data Index: 207568 (C4 (Web Text))

Max Activating Token Index: 985

Text #1

Max Range: 3.3938. Min Range: -3.3938

Max Act: 3.3624. Min Act: -0.1700

Data Index: 37918 (C4 (Web Text))

Max Activating Token Index: 814

Text #2

Max Range: 3.3938. Min Range: -3.3938

Max Act: 3.3781. Min Act: -0.1700

Data Index: 119649 (C4 (Web Text))

Max Activating Token Index: 953

Text #3

Max Range: 3.3938. Min Range: -3.3938

Max Act: 3.3467. Min Act: -0.1700

Data Index: 219592 (C4 (Web Text))

Max Activating Token Index: 177

Text #4

Max Range: 3.3938. Min Range: -3.3938

Max Act: 3.3467. Min Act: -0.1700

Data Index: 465086 (C4 (Web Text))

Max Activating Token Index: 926

Text #5

Max Range: 3.3938. Min Range: -3.3938

Max Act: 3.3467. Min Act: -0.1700

Data Index: 1337358 (C4 (Web Text))

Max Activating Token Index: 438

Text #6

Max Range: 3.3938. Min Range: -3.3938

Max Act: 3.2996. Min Act: -0.1700

Data Index: 1051090 (C4 (Web Text))

Max Activating Token Index: 523

Text #7

Max Range: 3.3938. Min Range: -3.3938

Max Act: 3.2839. Min Act: -0.1700

Data Index: 423795 (C4 (Web Text))

Max Activating Token Index: 970

Text #8

Max Range: 3.3938. Min Range: -3.3938

Max Act: 3.3310. Min Act: -0.1700

Data Index: 1056368 (C4 (Web Text))

Max Activating Token Index: 124

Text #9

Max Range: 3.3938. Min Range: -3.3938

Max Act: 3.3153. Min Act: -0.1700

Data Index: 495782 (C4 (Web Text))

Max Activating Token Index: 730

Text #10

Max Range: 3.3938. Min Range: -3.3938

Max Act: 3.3310. Min Act: -0.1700

Data Index: 570746 (C4 (Web Text))

Max Activating Token Index: 669

Text #11

Max Range: 3.3938. Min Range: -3.3938

Max Act: 3.2996. Min Act: -0.1700

Data Index: 1073684 (C4 (Web Text))

Max Activating Token Index: 986

Text #12

Max Range: 3.3938. Min Range: -3.3938

Max Act: 3.2839. Min Act: -0.1700

Data Index: 625850 (C4 (Web Text))

Max Activating Token Index: 495

Text #13

Max Range: 3.3938. Min Range: -3.3938

Max Act: 3.3153. Min Act: -0.1700

Data Index: 387106 (C4 (Web Text))

Max Activating Token Index: 568

Text #14

Max Range: 3.3938. Min Range: -3.3938

Max Act: 3.2996. Min Act: -0.1700

Data Index: 900091 (C4 (Web Text))

Max Activating Token Index: 700

Text #15

Max Range: 3.3938. Min Range: -3.3938

Max Act: 3.3153. Min Act: -0.1700

Data Index: 462912 (C4 (Web Text))

Max Activating Token Index: 538

Text #16

Max Range: 3.3938. Min Range: -3.3938

Max Act: 3.2996. Min Act: -0.1700

Data Index: 403216 (C4 (Web Text))

Max Activating Token Index: 314

Text #17

Max Range: 3.3938. Min Range: -3.3938

Max Act: 3.2839. Min Act: -0.1700

Data Index: 1199883 (C4 (Web Text))

Max Activating Token Index: 977

Text #18

Max Range: 3.3938. Min Range: -3.3938

Max Act: 3.2839. Min Act: -0.1700

Data Index: 219432 (C4 (Web Text))

Max Activating Token Index: 640

Text #19

Max Range: 3.3938. Min Range: -3.3938

Max Act: 3.2525. Min Act: -0.1700

Data Index: 322057 (C4 (Web Text))

Max Activating Token Index: 948