Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1493 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 3.1247. Min Range: -3.1247

Max Act: 3.1247. Min Act: -0.1700

Data Index: 6763 (C4 (Web Text))

Max Activating Token Index: 939

Text #1

Max Range: 3.1247. Min Range: -3.1247

Max Act: 3.1089. Min Act: -0.1700

Data Index: 152901 (C4 (Web Text))

Max Activating Token Index: 586

Text #2

Max Range: 3.1247. Min Range: -3.1247

Max Act: 3.0932. Min Act: -0.1700

Data Index: 416207 (C4 (Web Text))

Max Activating Token Index: 930

Text #3

Max Range: 3.1247. Min Range: -3.1247

Max Act: 3.0932. Min Act: -0.1700

Data Index: 952514 (C4 (Web Text))

Max Activating Token Index: 163

Text #4

Max Range: 3.1247. Min Range: -3.1247

Max Act: 3.0932. Min Act: -0.1700

Data Index: 1320642 (C4 (Web Text))

Max Activating Token Index: 219

Text #5

Max Range: 3.1247. Min Range: -3.1247

Max Act: 3.0774. Min Act: -0.1700

Data Index: 610712 (C4 (Web Text))

Max Activating Token Index: 924

Text #6

Max Range: 3.1247. Min Range: -3.1247

Max Act: 3.0774. Min Act: -0.1700

Data Index: 675272 (C4 (Web Text))

Max Activating Token Index: 434

Text #7

Max Range: 3.1247. Min Range: -3.1247

Max Act: 3.0932. Min Act: -0.1700

Data Index: 322114 (C4 (Web Text))

Max Activating Token Index: 941

Text #8

Max Range: 3.1247. Min Range: -3.1247

Max Act: 3.0774. Min Act: -0.1700

Data Index: 1293976 (C4 (Web Text))

Max Activating Token Index: 340

Text #9

Max Range: 3.1247. Min Range: -3.1247

Max Act: 3.0458. Min Act: -0.1700

Data Index: 355930 (C4 (Web Text))

Max Activating Token Index: 168

Text #10

Max Range: 3.1247. Min Range: -3.1247

Max Act: 3.0458. Min Act: -0.1700

Data Index: 1638625 (Python Code)

Max Activating Token Index: 717

Text #11

Max Range: 3.1247. Min Range: -3.1247

Max Act: 3.0458. Min Act: -0.1700

Data Index: 1524481 (Python Code)

Max Activating Token Index: 938

Text #12

Max Range: 3.1247. Min Range: -3.1247

Max Act: 3.0458. Min Act: -0.1700

Data Index: 729265 (C4 (Web Text))

Max Activating Token Index: 217

Text #13

Max Range: 3.1247. Min Range: -3.1247

Max Act: 3.0458. Min Act: -0.1700

Data Index: 616676 (C4 (Web Text))

Max Activating Token Index: 399

Text #14

Max Range: 3.1247. Min Range: -3.1247

Max Act: 3.0458. Min Act: -0.1700

Data Index: 982086 (C4 (Web Text))

Max Activating Token Index: 464

Text #15

Max Range: 3.1247. Min Range: -3.1247

Max Act: 3.0458. Min Act: -0.1700

Data Index: 765593 (C4 (Web Text))

Max Activating Token Index: 276

Text #16

Max Range: 3.1247. Min Range: -3.1247

Max Act: 3.0458. Min Act: -0.1700

Data Index: 222288 (C4 (Web Text))

Max Activating Token Index: 740

Text #17

Max Range: 3.1247. Min Range: -3.1247

Max Act: 3.0458. Min Act: -0.1700

Data Index: 418316 (C4 (Web Text))

Max Activating Token Index: 323

Text #18

Max Range: 3.1247. Min Range: -3.1247

Max Act: 3.0458. Min Act: -0.1700

Data Index: 666506 (C4 (Web Text))

Max Activating Token Index: 268

Text #19

Max Range: 3.1247. Min Range: -3.1247

Max Act: 3.0458. Min Act: -0.1700

Data Index: 121154 (C4 (Web Text))

Max Activating Token Index: 99