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

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

Neuron 814 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: 5.4315. Min Range: -5.4315

Max Act: 5.4315. Min Act: -0.1700

Data Index: 770492 (C4 (Web Text))

Max Activating Token Index: 426

Text #1

Max Range: 5.4315. Min Range: -5.4315

Max Act: 5.2440. Min Act: -0.1700

Data Index: 616305 (C4 (Web Text))

Max Activating Token Index: 239

Text #2

Max Range: 5.4315. Min Range: -5.4315

Max Act: 5.1815. Min Act: -0.1700

Data Index: 564498 (C4 (Web Text))

Max Activating Token Index: 186

Text #3

Max Range: 5.4315. Min Range: -5.4315

Max Act: 5.0565. Min Act: -0.1700

Data Index: 891782 (C4 (Web Text))

Max Activating Token Index: 166

Text #4

Max Range: 5.4315. Min Range: -5.4315

Max Act: 4.9940. Min Act: -0.1700

Data Index: 335929 (C4 (Web Text))

Max Activating Token Index: 238

Text #5

Max Range: 5.4315. Min Range: -5.4315

Max Act: 5.1190. Min Act: -0.1700

Data Index: 611715 (C4 (Web Text))

Max Activating Token Index: 124

Text #6

Max Range: 5.4315. Min Range: -5.4315

Max Act: 5.0565. Min Act: -0.1700

Data Index: 846254 (C4 (Web Text))

Max Activating Token Index: 334

Text #7

Max Range: 5.4315. Min Range: -5.4315

Max Act: 5.0565. Min Act: -0.1700

Data Index: 673428 (C4 (Web Text))

Max Activating Token Index: 232

Text #8

Max Range: 5.4315. Min Range: -5.4315

Max Act: 4.9940. Min Act: -0.1700

Data Index: 206051 (C4 (Web Text))

Max Activating Token Index: 231

Text #9

Max Range: 5.4315. Min Range: -5.4315

Max Act: 4.9940. Min Act: -0.1699

Data Index: 327436 (C4 (Web Text))

Max Activating Token Index: 266

Text #10

Max Range: 5.4315. Min Range: -5.4315

Max Act: 4.9315. Min Act: -0.1700

Data Index: 116635 (C4 (Web Text))

Max Activating Token Index: 883

Text #11

Max Range: 5.4315. Min Range: -5.4315

Max Act: 4.9315. Min Act: -0.1700

Data Index: 78810 (C4 (Web Text))

Max Activating Token Index: 874

Text #12

Max Range: 5.4315. Min Range: -5.4315

Max Act: 4.9315. Min Act: -0.1700

Data Index: 975852 (C4 (Web Text))

Max Activating Token Index: 985

Text #13

Max Range: 5.4315. Min Range: -5.4315

Max Act: 4.9315. Min Act: -0.1700

Data Index: 166069 (C4 (Web Text))

Max Activating Token Index: 136

Text #14

Max Range: 5.4315. Min Range: -5.4315

Max Act: 4.9940. Min Act: -0.1699

Data Index: 296238 (C4 (Web Text))

Max Activating Token Index: 758

Text #15

Max Range: 5.4315. Min Range: -5.4315

Max Act: 4.9315. Min Act: -0.1700

Data Index: 594329 (C4 (Web Text))

Max Activating Token Index: 480

Text #16

Max Range: 5.4315. Min Range: -5.4315

Max Act: 4.9315. Min Act: -0.1700

Data Index: 13883 (C4 (Web Text))

Max Activating Token Index: 584

Text #17

Max Range: 5.4315. Min Range: -5.4315

Max Act: 4.8690. Min Act: -0.1700

Data Index: 1443373 (Python Code)

Max Activating Token Index: 166

Text #18

Max Range: 5.4315. Min Range: -5.4315

Max Act: 4.8690. Min Act: -0.1700

Data Index: 1466020 (Python Code)

Max Activating Token Index: 361

Text #19

Max Range: 5.4315. Min Range: -5.4315

Max Act: 4.8690. Min Act: -0.1700

Data Index: 1422322 (Python Code)

Max Activating Token Index: 619