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

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

Neuron 646 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: 3.9181. Min Range: -3.9181

Max Act: 3.9181. Min Act: -0.1700

Data Index: 473129 (C4 (Web Text))

Max Activating Token Index: 641

Text #1

Max Range: 3.9181. Min Range: -3.9181

Max Act: 3.8399. Min Act: -0.1700

Data Index: 66992 (C4 (Web Text))

Max Activating Token Index: 357

Text #2

Max Range: 3.9181. Min Range: -3.9181

Max Act: 3.7773. Min Act: -0.1700

Data Index: 169673 (C4 (Web Text))

Max Activating Token Index: 926

Text #3

Max Range: 3.9181. Min Range: -3.9181

Max Act: 3.7617. Min Act: -0.1699

Data Index: 1445169 (Python Code)

Max Activating Token Index: 143

Text #4

Max Range: 3.9181. Min Range: -3.9181

Max Act: 3.7617. Min Act: -0.1700

Data Index: 1044837 (C4 (Web Text))

Max Activating Token Index: 239

Text #5

Max Range: 3.9181. Min Range: -3.9181

Max Act: 3.6835. Min Act: -0.1700

Data Index: 1390476 (Python Code)

Max Activating Token Index: 849

Text #6

Max Range: 3.9181. Min Range: -3.9181

Max Act: 3.6835. Min Act: -0.1700

Data Index: 763760 (C4 (Web Text))

Max Activating Token Index: 534

Text #7

Max Range: 3.9181. Min Range: -3.9181

Max Act: 3.6835. Min Act: -0.1700

Data Index: 1110618 (C4 (Web Text))

Max Activating Token Index: 515

Text #8

Max Range: 3.9181. Min Range: -3.9181

Max Act: 3.6678. Min Act: -0.1700

Data Index: 1501431 (Python Code)

Max Activating Token Index: 847

Text #9

Max Range: 3.9181. Min Range: -3.9181

Max Act: 3.6835. Min Act: -0.1700

Data Index: 1214651 (C4 (Web Text))

Max Activating Token Index: 480

Text #10

Max Range: 3.9181. Min Range: -3.9181

Max Act: 3.6678. Min Act: -0.1700

Data Index: 713180 (C4 (Web Text))

Max Activating Token Index: 322

Text #11

Max Range: 3.9181. Min Range: -3.9181

Max Act: 3.6678. Min Act: -0.1700

Data Index: 200834 (C4 (Web Text))

Max Activating Token Index: 253

Text #12

Max Range: 3.9181. Min Range: -3.9181

Max Act: 3.6835. Min Act: -0.1700

Data Index: 857254 (C4 (Web Text))

Max Activating Token Index: 449

Text #13

Max Range: 3.9181. Min Range: -3.9181

Max Act: 3.6835. Min Act: -0.1700

Data Index: 22572 (C4 (Web Text))

Max Activating Token Index: 301

Text #14

Max Range: 3.9181. Min Range: -3.9181

Max Act: 3.6209. Min Act: -0.1700

Data Index: 715597 (C4 (Web Text))

Max Activating Token Index: 478

Text #15

Max Range: 3.9181. Min Range: -3.9181

Max Act: 3.6209. Min Act: -0.1700

Data Index: 1011360 (C4 (Web Text))

Max Activating Token Index: 520

Text #16

Max Range: 3.9181. Min Range: -3.9181

Max Act: 3.6365. Min Act: -0.1700

Data Index: 570686 (C4 (Web Text))

Max Activating Token Index: 315

Text #17

Max Range: 3.9181. Min Range: -3.9181

Max Act: 3.5896. Min Act: -0.1700

Data Index: 102649 (C4 (Web Text))

Max Activating Token Index: 912

Text #18

Max Range: 3.9181. Min Range: -3.9181

Max Act: 3.6522. Min Act: -0.1700

Data Index: 441072 (C4 (Web Text))

Max Activating Token Index: 713

Text #19

Max Range: 3.9181. Min Range: -3.9181

Max Act: 3.5739. Min Act: -0.1700

Data Index: 821733 (C4 (Web Text))

Max Activating Token Index: 737