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

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

Neuron 1605 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.6461. Min Range: -5.6461

Max Act: 5.6461. Min Act: -0.1700

Data Index: 320290 (C4 (Web Text))

Max Activating Token Index: 56

Text #1

Max Range: 5.6461. Min Range: -5.6461

Max Act: 5.5211. Min Act: -0.1700

Data Index: 866129 (C4 (Web Text))

Max Activating Token Index: 188

Text #2

Max Range: 5.6461. Min Range: -5.6461

Max Act: 5.3023. Min Act: -0.1700

Data Index: 1341927 (C4 (Web Text))

Max Activating Token Index: 325

Text #3

Max Range: 5.6461. Min Range: -5.6461

Max Act: 5.2398. Min Act: -0.1700

Data Index: 1236238 (C4 (Web Text))

Max Activating Token Index: 481

Text #4

Max Range: 5.6461. Min Range: -5.6461

Max Act: 5.2398. Min Act: -0.1700

Data Index: 418651 (C4 (Web Text))

Max Activating Token Index: 324

Text #5

Max Range: 5.6461. Min Range: -5.6461

Max Act: 5.0836. Min Act: -0.1700

Data Index: 1111286 (C4 (Web Text))

Max Activating Token Index: 138

Text #6

Max Range: 5.6461. Min Range: -5.6461

Max Act: 4.9898. Min Act: -0.1700

Data Index: 1215944 (C4 (Web Text))

Max Activating Token Index: 852

Text #7

Max Range: 5.6461. Min Range: -5.6461

Max Act: 5.0211. Min Act: -0.1700

Data Index: 696687 (C4 (Web Text))

Max Activating Token Index: 136

Text #8

Max Range: 5.6461. Min Range: -5.6461

Max Act: 4.8961. Min Act: -0.1700

Data Index: 661678 (C4 (Web Text))

Max Activating Token Index: 847

Text #9

Max Range: 5.6461. Min Range: -5.6461

Max Act: 4.9273. Min Act: -0.1700

Data Index: 860386 (C4 (Web Text))

Max Activating Token Index: 324

Text #10

Max Range: 5.6461. Min Range: -5.6461

Max Act: 4.9586. Min Act: -0.1700

Data Index: 358573 (C4 (Web Text))

Max Activating Token Index: 449

Text #11

Max Range: 5.6461. Min Range: -5.6461

Max Act: 4.9273. Min Act: -0.1700

Data Index: 557818 (C4 (Web Text))

Max Activating Token Index: 812

Text #12

Max Range: 5.6461. Min Range: -5.6461

Max Act: 4.8961. Min Act: -0.1700

Data Index: 1025270 (C4 (Web Text))

Max Activating Token Index: 800

Text #13

Max Range: 5.6461. Min Range: -5.6461

Max Act: 4.8648. Min Act: -0.1700

Data Index: 508614 (C4 (Web Text))

Max Activating Token Index: 485

Text #14

Max Range: 5.6461. Min Range: -5.6461

Max Act: 4.8336. Min Act: -0.1700

Data Index: 197114 (C4 (Web Text))

Max Activating Token Index: 503

Text #15

Max Range: 5.6461. Min Range: -5.6461

Max Act: 4.7398. Min Act: -0.1700

Data Index: 14729 (C4 (Web Text))

Max Activating Token Index: 813

Text #16

Max Range: 5.6461. Min Range: -5.6461

Max Act: 4.7711. Min Act: -0.1700

Data Index: 513150 (C4 (Web Text))

Max Activating Token Index: 385

Text #17

Max Range: 5.6461. Min Range: -5.6461

Max Act: 4.7398. Min Act: -0.1700

Data Index: 144254 (C4 (Web Text))

Max Activating Token Index: 176

Text #18

Max Range: 5.6461. Min Range: -5.6461

Max Act: 4.7711. Min Act: -0.1700

Data Index: 1198348 (C4 (Web Text))

Max Activating Token Index: 174

Text #19

Max Range: 5.6461. Min Range: -5.6461

Max Act: 4.7398. Min Act: -0.1700

Data Index: 996152 (C4 (Web Text))

Max Activating Token Index: 851