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

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

Neuron 2002 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 3.5446. Min Range: -3.5446

Max Act: 3.5446. Min Act: -0.1700

Data Index: 1140644 (C4 (Web Text))

Max Activating Token Index: 666

Text #1

Max Range: 3.5446. Min Range: -3.5446

Max Act: 3.3879. Min Act: -0.1700

Data Index: 556190 (C4 (Web Text))

Max Activating Token Index: 443

Text #2

Max Range: 3.5446. Min Range: -3.5446

Max Act: 3.3879. Min Act: -0.1700

Data Index: 58389 (C4 (Web Text))

Max Activating Token Index: 566

Text #3

Max Range: 3.5446. Min Range: -3.5446

Max Act: 3.3565. Min Act: -0.1700

Data Index: 905677 (C4 (Web Text))

Max Activating Token Index: 562

Text #4

Max Range: 3.5446. Min Range: -3.5446

Max Act: 3.3879. Min Act: -0.1700

Data Index: 55406 (C4 (Web Text))

Max Activating Token Index: 257

Text #5

Max Range: 3.5446. Min Range: -3.5446

Max Act: 3.3565. Min Act: -0.1700

Data Index: 427934 (C4 (Web Text))

Max Activating Token Index: 174

Text #6

Max Range: 3.5446. Min Range: -3.5446

Max Act: 3.3722. Min Act: -0.1700

Data Index: 1284753 (C4 (Web Text))

Max Activating Token Index: 812

Text #7

Max Range: 3.5446. Min Range: -3.5446

Max Act: 3.3408. Min Act: -0.1700

Data Index: 351450 (C4 (Web Text))

Max Activating Token Index: 885

Text #8

Max Range: 3.5446. Min Range: -3.5446

Max Act: 3.3408. Min Act: -0.1700

Data Index: 418495 (C4 (Web Text))

Max Activating Token Index: 1005

Text #9

Max Range: 3.5446. Min Range: -3.5446

Max Act: 3.3251. Min Act: -0.1700

Data Index: 168235 (C4 (Web Text))

Max Activating Token Index: 587

Text #10

Max Range: 3.5446. Min Range: -3.5446

Max Act: 3.3094. Min Act: -0.1700

Data Index: 458428 (C4 (Web Text))

Max Activating Token Index: 328

Text #11

Max Range: 3.5446. Min Range: -3.5446

Max Act: 3.3094. Min Act: -0.1700

Data Index: 699379 (C4 (Web Text))

Max Activating Token Index: 641

Text #12

Max Range: 3.5446. Min Range: -3.5446

Max Act: 3.3094. Min Act: -0.1700

Data Index: 949595 (C4 (Web Text))

Max Activating Token Index: 237

Text #13

Max Range: 3.5446. Min Range: -3.5446

Max Act: 3.2937. Min Act: -0.1700

Data Index: 58388 (C4 (Web Text))

Max Activating Token Index: 615

Text #14

Max Range: 3.5446. Min Range: -3.5446

Max Act: 3.3094. Min Act: -0.1700

Data Index: 1096257 (C4 (Web Text))

Max Activating Token Index: 511

Text #15

Max Range: 3.5446. Min Range: -3.5446

Max Act: 3.2623. Min Act: -0.1700

Data Index: 965958 (C4 (Web Text))

Max Activating Token Index: 939

Text #16

Max Range: 3.5446. Min Range: -3.5446

Max Act: 3.2780. Min Act: -0.1700

Data Index: 537979 (C4 (Web Text))

Max Activating Token Index: 655

Text #17

Max Range: 3.5446. Min Range: -3.5446

Max Act: 3.2780. Min Act: -0.1700

Data Index: 107247 (C4 (Web Text))

Max Activating Token Index: 316

Text #18

Max Range: 3.5446. Min Range: -3.5446

Max Act: 3.2780. Min Act: -0.1700

Data Index: 1128868 (C4 (Web Text))

Max Activating Token Index: 644

Text #19

Max Range: 3.5446. Min Range: -3.5446

Max Act: 3.2466. Min Act: -0.1700

Data Index: 495251 (C4 (Web Text))

Max Activating Token Index: 668