Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 755 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 5.4240. Min Range: -5.4240

Max Act: 5.4240. Min Act: -0.1700

Data Index: 48390 (C4 (Web Text))

Max Activating Token Index: 27

Text #1

Max Range: 5.4240. Min Range: -5.4240

Max Act: 5.3928. Min Act: -0.1700

Data Index: 1310539 (C4 (Web Text))

Max Activating Token Index: 555

Text #2

Max Range: 5.4240. Min Range: -5.4240

Max Act: 5.0803. Min Act: -0.1700

Data Index: 289302 (C4 (Web Text))

Max Activating Token Index: 88

Text #3

Max Range: 5.4240. Min Range: -5.4240

Max Act: 5.0178. Min Act: -0.1700

Data Index: 929613 (C4 (Web Text))

Max Activating Token Index: 584

Text #4

Max Range: 5.4240. Min Range: -5.4240

Max Act: 5.0490. Min Act: -0.1700

Data Index: 741938 (C4 (Web Text))

Max Activating Token Index: 45

Text #5

Max Range: 5.4240. Min Range: -5.4240

Max Act: 5.0803. Min Act: -0.1700

Data Index: 630939 (C4 (Web Text))

Max Activating Token Index: 166

Text #6

Max Range: 5.4240. Min Range: -5.4240

Max Act: 5.0490. Min Act: -0.1700

Data Index: 1032742 (C4 (Web Text))

Max Activating Token Index: 636

Text #7

Max Range: 5.4240. Min Range: -5.4240

Max Act: 5.0490. Min Act: -0.1700

Data Index: 1583937 (Python Code)

Max Activating Token Index: 61

Text #8

Max Range: 5.4240. Min Range: -5.4240

Max Act: 5.0803. Min Act: -0.1700

Data Index: 664178 (C4 (Web Text))

Max Activating Token Index: 159

Text #9

Max Range: 5.4240. Min Range: -5.4240

Max Act: 4.9553. Min Act: -0.1700

Data Index: 973830 (C4 (Web Text))

Max Activating Token Index: 26

Text #10

Max Range: 5.4240. Min Range: -5.4240

Max Act: 5.0803. Min Act: -0.1700

Data Index: 818846 (C4 (Web Text))

Max Activating Token Index: 546

Text #11

Max Range: 5.4240. Min Range: -5.4240

Max Act: 4.9865. Min Act: -0.1700

Data Index: 1508935 (Python Code)

Max Activating Token Index: 108

Text #12

Max Range: 5.4240. Min Range: -5.4240

Max Act: 4.9865. Min Act: -0.1700

Data Index: 1533994 (Python Code)

Max Activating Token Index: 108

Text #13

Max Range: 5.4240. Min Range: -5.4240

Max Act: 5.0178. Min Act: -0.1700

Data Index: 1489338 (Python Code)

Max Activating Token Index: 66

Text #14

Max Range: 5.4240. Min Range: -5.4240

Max Act: 4.9865. Min Act: -0.1700

Data Index: 1620511 (Python Code)

Max Activating Token Index: 93

Text #15

Max Range: 5.4240. Min Range: -5.4240

Max Act: 4.9240. Min Act: -0.1700

Data Index: 1585554 (Python Code)

Max Activating Token Index: 71

Text #16

Max Range: 5.4240. Min Range: -5.4240

Max Act: 5.0490. Min Act: -0.1700

Data Index: 1079724 (C4 (Web Text))

Max Activating Token Index: 274

Text #17

Max Range: 5.4240. Min Range: -5.4240

Max Act: 5.0178. Min Act: -0.1700

Data Index: 66740 (C4 (Web Text))

Max Activating Token Index: 249

Text #18

Max Range: 5.4240. Min Range: -5.4240

Max Act: 5.0178. Min Act: -0.1700

Data Index: 404328 (C4 (Web Text))

Max Activating Token Index: 625

Text #19

Max Range: 5.4240. Min Range: -5.4240

Max Act: 4.8303. Min Act: -0.1700

Data Index: 311618 (C4 (Web Text))

Max Activating Token Index: 97