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

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

Neuron 8 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: 3.4480. Min Range: -3.4480

Max Act: 3.4480. Min Act: -0.1700

Data Index: 1246356 (C4 (Web Text))

Max Activating Token Index: 241

Text #1

Max Range: 3.4480. Min Range: -3.4480

Max Act: 3.3382. Min Act: -0.1700

Data Index: 908805 (C4 (Web Text))

Max Activating Token Index: 469

Text #2

Max Range: 3.4480. Min Range: -3.4480

Max Act: 3.3225. Min Act: -0.1700

Data Index: 768902 (C4 (Web Text))

Max Activating Token Index: 583

Text #3

Max Range: 3.4480. Min Range: -3.4480

Max Act: 3.2911. Min Act: -0.1700

Data Index: 969971 (C4 (Web Text))

Max Activating Token Index: 30

Text #4

Max Range: 3.4480. Min Range: -3.4480

Max Act: 3.2754. Min Act: -0.1700

Data Index: 1015357 (C4 (Web Text))

Max Activating Token Index: 340

Text #5

Max Range: 3.4480. Min Range: -3.4480

Max Act: 3.2440. Min Act: -0.1700

Data Index: 216679 (C4 (Web Text))

Max Activating Token Index: 958

Text #6

Max Range: 3.4480. Min Range: -3.4480

Max Act: 3.2282. Min Act: -0.1700

Data Index: 472178 (C4 (Web Text))

Max Activating Token Index: 795

Text #7

Max Range: 3.4480. Min Range: -3.4480

Max Act: 3.2440. Min Act: -0.1700

Data Index: 1043761 (C4 (Web Text))

Max Activating Token Index: 166

Text #8

Max Range: 3.4480. Min Range: -3.4480

Max Act: 3.2282. Min Act: -0.1700

Data Index: 707083 (C4 (Web Text))

Max Activating Token Index: 85

Text #9

Max Range: 3.4480. Min Range: -3.4480

Max Act: 3.2125. Min Act: -0.1700

Data Index: 1230489 (C4 (Web Text))

Max Activating Token Index: 233

Text #10

Max Range: 3.4480. Min Range: -3.4480

Max Act: 3.2282. Min Act: -0.1700

Data Index: 300167 (C4 (Web Text))

Max Activating Token Index: 171

Text #11

Max Range: 3.4480. Min Range: -3.4480

Max Act: 3.1810. Min Act: -0.1700

Data Index: 80488 (C4 (Web Text))

Max Activating Token Index: 972

Text #12

Max Range: 3.4480. Min Range: -3.4480

Max Act: 3.2125. Min Act: -0.1700

Data Index: 1349550 (C4 (Web Text))

Max Activating Token Index: 194

Text #13

Max Range: 3.4480. Min Range: -3.4480

Max Act: 3.1968. Min Act: -0.1700

Data Index: 1312327 (C4 (Web Text))

Max Activating Token Index: 52

Text #14

Max Range: 3.4480. Min Range: -3.4480

Max Act: 3.1810. Min Act: -0.1700

Data Index: 780959 (C4 (Web Text))

Max Activating Token Index: 199

Text #15

Max Range: 3.4480. Min Range: -3.4480

Max Act: 3.1653. Min Act: -0.1700

Data Index: 1279857 (C4 (Web Text))

Max Activating Token Index: 383

Text #16

Max Range: 3.4480. Min Range: -3.4480

Max Act: 3.1496. Min Act: -0.1700

Data Index: 1114136 (C4 (Web Text))

Max Activating Token Index: 41

Text #17

Max Range: 3.4480. Min Range: -3.4480

Max Act: 3.1496. Min Act: -0.1700

Data Index: 484263 (C4 (Web Text))

Max Activating Token Index: 852

Text #18

Max Range: 3.4480. Min Range: -3.4480

Max Act: 3.1653. Min Act: -0.1700

Data Index: 751012 (C4 (Web Text))

Max Activating Token Index: 254

Text #19

Max Range: 3.4480. Min Range: -3.4480

Max Act: 3.1496. Min Act: -0.1700

Data Index: 769320 (C4 (Web Text))

Max Activating Token Index: 845