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

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

Neuron 1024 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: 5.7864. Min Range: -5.7864

Max Act: 5.7864. Min Act: -0.1700

Data Index: 1330815 (C4 (Web Text))

Max Activating Token Index: 671

Text #1

Max Range: 5.7864. Min Range: -5.7864

Max Act: 5.5364. Min Act: -0.1700

Data Index: 578253 (C4 (Web Text))

Max Activating Token Index: 873

Text #2

Max Range: 5.7864. Min Range: -5.7864

Max Act: 5.5364. Min Act: -0.1700

Data Index: 917123 (C4 (Web Text))

Max Activating Token Index: 962

Text #3

Max Range: 5.7864. Min Range: -5.7864

Max Act: 5.5052. Min Act: -0.1700

Data Index: 387602 (C4 (Web Text))

Max Activating Token Index: 733

Text #4

Max Range: 5.7864. Min Range: -5.7864

Max Act: 5.6614. Min Act: -0.1700

Data Index: 231748 (C4 (Web Text))

Max Activating Token Index: 593

Text #5

Max Range: 5.7864. Min Range: -5.7864

Max Act: 5.4427. Min Act: -0.1700

Data Index: 1093745 (C4 (Web Text))

Max Activating Token Index: 412

Text #6

Max Range: 5.7864. Min Range: -5.7864

Max Act: 5.5364. Min Act: -0.1700

Data Index: 903799 (C4 (Web Text))

Max Activating Token Index: 81

Text #7

Max Range: 5.7864. Min Range: -5.7864

Max Act: 5.5364. Min Act: -0.1700

Data Index: 483722 (C4 (Web Text))

Max Activating Token Index: 864

Text #8

Max Range: 5.7864. Min Range: -5.7864

Max Act: 5.4739. Min Act: -0.1700

Data Index: 630545 (C4 (Web Text))

Max Activating Token Index: 176

Text #9

Max Range: 5.7864. Min Range: -5.7864

Max Act: 5.4114. Min Act: -0.1700

Data Index: 175462 (C4 (Web Text))

Max Activating Token Index: 775

Text #10

Max Range: 5.7864. Min Range: -5.7864

Max Act: 5.5052. Min Act: -0.1700

Data Index: 428867 (C4 (Web Text))

Max Activating Token Index: 402

Text #11

Max Range: 5.7864. Min Range: -5.7864

Max Act: 5.4427. Min Act: -0.1700

Data Index: 1384019 (Python Code)

Max Activating Token Index: 460

Text #12

Max Range: 5.7864. Min Range: -5.7864

Max Act: 5.4114. Min Act: -0.1700

Data Index: 366464 (C4 (Web Text))

Max Activating Token Index: 263

Text #13

Max Range: 5.7864. Min Range: -5.7864

Max Act: 5.4739. Min Act: -0.1700

Data Index: 612899 (C4 (Web Text))

Max Activating Token Index: 305

Text #14

Max Range: 5.7864. Min Range: -5.7864

Max Act: 5.4427. Min Act: -0.1700

Data Index: 141906 (C4 (Web Text))

Max Activating Token Index: 726

Text #15

Max Range: 5.7864. Min Range: -5.7864

Max Act: 5.4427. Min Act: -0.1700

Data Index: 7536 (C4 (Web Text))

Max Activating Token Index: 591

Text #16

Max Range: 5.7864. Min Range: -5.7864

Max Act: 5.4114. Min Act: -0.1700

Data Index: 1335951 (C4 (Web Text))

Max Activating Token Index: 228

Text #17

Max Range: 5.7864. Min Range: -5.7864

Max Act: 5.3489. Min Act: -0.1700

Data Index: 1337552 (C4 (Web Text))

Max Activating Token Index: 951

Text #18

Max Range: 5.7864. Min Range: -5.7864

Max Act: 5.4114. Min Act: -0.1700

Data Index: 295454 (C4 (Web Text))

Max Activating Token Index: 677

Text #19

Max Range: 5.7864. Min Range: -5.7864

Max Act: 5.4114. Min Act: -0.1700

Data Index: 1013112 (C4 (Web Text))

Max Activating Token Index: 939