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

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

Neuron 1600 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: 3.7647. Min Range: -3.7647

Max Act: 3.7647. Min Act: -0.1700

Data Index: 1337228 (C4 (Web Text))

Max Activating Token Index: 316

Text #1

Max Range: 3.7647. Min Range: -3.7647

Max Act: 3.7491. Min Act: -0.1700

Data Index: 350974 (C4 (Web Text))

Max Activating Token Index: 750

Text #2

Max Range: 3.7647. Min Range: -3.7647

Max Act: 3.6552. Min Act: -0.1700

Data Index: 388103 (C4 (Web Text))

Max Activating Token Index: 719

Text #3

Max Range: 3.7647. Min Range: -3.7647

Max Act: 3.5926. Min Act: -0.1700

Data Index: 956841 (C4 (Web Text))

Max Activating Token Index: 502

Text #4

Max Range: 3.7647. Min Range: -3.7647

Max Act: 3.5456. Min Act: -0.1700

Data Index: 528171 (C4 (Web Text))

Max Activating Token Index: 257

Text #5

Max Range: 3.7647. Min Range: -3.7647

Max Act: 3.5142. Min Act: -0.1700

Data Index: 952051 (C4 (Web Text))

Max Activating Token Index: 170

Text #6

Max Range: 3.7647. Min Range: -3.7647

Max Act: 3.4829. Min Act: -0.1700

Data Index: 164601 (C4 (Web Text))

Max Activating Token Index: 863

Text #7

Max Range: 3.7647. Min Range: -3.7647

Max Act: 3.4672. Min Act: -0.1700

Data Index: 1046931 (C4 (Web Text))

Max Activating Token Index: 406

Text #8

Max Range: 3.7647. Min Range: -3.7647

Max Act: 3.4202. Min Act: -0.1700

Data Index: 829338 (C4 (Web Text))

Max Activating Token Index: 491

Text #9

Max Range: 3.7647. Min Range: -3.7647

Max Act: 3.4045. Min Act: -0.1700

Data Index: 1244014 (C4 (Web Text))

Max Activating Token Index: 692

Text #10

Max Range: 3.7647. Min Range: -3.7647

Max Act: 3.3731. Min Act: -0.1700

Data Index: 881922 (C4 (Web Text))

Max Activating Token Index: 924

Text #11

Max Range: 3.7647. Min Range: -3.7647

Max Act: 3.3104. Min Act: -0.1700

Data Index: 379927 (C4 (Web Text))

Max Activating Token Index: 277

Text #12

Max Range: 3.7647. Min Range: -3.7647

Max Act: 3.3575. Min Act: -0.1700

Data Index: 72649 (C4 (Web Text))

Max Activating Token Index: 851

Text #13

Max Range: 3.7647. Min Range: -3.7647

Max Act: 3.3418. Min Act: -0.1700

Data Index: 1314142 (C4 (Web Text))

Max Activating Token Index: 70

Text #14

Max Range: 3.7647. Min Range: -3.7647

Max Act: 3.2789. Min Act: -0.1700

Data Index: 399545 (C4 (Web Text))

Max Activating Token Index: 391

Text #15

Max Range: 3.7647. Min Range: -3.7647

Max Act: 3.3104. Min Act: -0.1700

Data Index: 158445 (C4 (Web Text))

Max Activating Token Index: 946

Text #16

Max Range: 3.7647. Min Range: -3.7647

Max Act: 3.3261. Min Act: -0.1700

Data Index: 852228 (C4 (Web Text))

Max Activating Token Index: 292

Text #17

Max Range: 3.7647. Min Range: -3.7647

Max Act: 3.3418. Min Act: -0.1700

Data Index: 1045685 (C4 (Web Text))

Max Activating Token Index: 412

Text #18

Max Range: 3.7647. Min Range: -3.7647

Max Act: 3.3104. Min Act: -0.1700

Data Index: 912065 (C4 (Web Text))

Max Activating Token Index: 859

Text #19

Max Range: 3.7647. Min Range: -3.7647

Max Act: 3.3104. Min Act: -0.1700

Data Index: 155720 (C4 (Web Text))

Max Activating Token Index: 497