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

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

Neuron 804 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.4062. Min Range: -3.4062

Max Act: 3.4062. Min Act: -0.1700

Data Index: 908091 (C4 (Web Text))

Max Activating Token Index: 608

Text #1

Max Range: 3.4062. Min Range: -3.4062

Max Act: 3.3435. Min Act: -0.1700

Data Index: 457165 (C4 (Web Text))

Max Activating Token Index: 687

Text #2

Max Range: 3.4062. Min Range: -3.4062

Max Act: 3.3121. Min Act: -0.1700

Data Index: 274316 (C4 (Web Text))

Max Activating Token Index: 333

Text #3

Max Range: 3.4062. Min Range: -3.4062

Max Act: 3.2964. Min Act: -0.1700

Data Index: 678818 (C4 (Web Text))

Max Activating Token Index: 351

Text #4

Max Range: 3.4062. Min Range: -3.4062

Max Act: 3.3121. Min Act: -0.1700

Data Index: 340690 (C4 (Web Text))

Max Activating Token Index: 541

Text #5

Max Range: 3.4062. Min Range: -3.4062

Max Act: 3.2806. Min Act: -0.1700

Data Index: 988396 (C4 (Web Text))

Max Activating Token Index: 649

Text #6

Max Range: 3.4062. Min Range: -3.4062

Max Act: 3.2806. Min Act: -0.1700

Data Index: 1509564 (Python Code)

Max Activating Token Index: 991

Text #7

Max Range: 3.4062. Min Range: -3.4062

Max Act: 3.2649. Min Act: -0.1700

Data Index: 201952 (C4 (Web Text))

Max Activating Token Index: 844

Text #8

Max Range: 3.4062. Min Range: -3.4062

Max Act: 3.2492. Min Act: -0.1700

Data Index: 1351749 (C4 (Web Text))

Max Activating Token Index: 518

Text #9

Max Range: 3.4062. Min Range: -3.4062

Max Act: 3.2806. Min Act: -0.1700

Data Index: 1329951 (C4 (Web Text))

Max Activating Token Index: 210

Text #10

Max Range: 3.4062. Min Range: -3.4062

Max Act: 3.2178. Min Act: -0.1700

Data Index: 235321 (C4 (Web Text))

Max Activating Token Index: 697

Text #11

Max Range: 3.4062. Min Range: -3.4062

Max Act: 3.2335. Min Act: -0.1700

Data Index: 785236 (C4 (Web Text))

Max Activating Token Index: 22

Text #12

Max Range: 3.4062. Min Range: -3.4062

Max Act: 3.2178. Min Act: -0.1700

Data Index: 915137 (C4 (Web Text))

Max Activating Token Index: 672

Text #13

Max Range: 3.4062. Min Range: -3.4062

Max Act: 3.2335. Min Act: -0.1700

Data Index: 596735 (C4 (Web Text))

Max Activating Token Index: 1023

Text #14

Max Range: 3.4062. Min Range: -3.4062

Max Act: 3.2020. Min Act: -0.1700

Data Index: 1282026 (C4 (Web Text))

Max Activating Token Index: 546

Text #15

Max Range: 3.4062. Min Range: -3.4062

Max Act: 3.2178. Min Act: -0.1700

Data Index: 483645 (C4 (Web Text))

Max Activating Token Index: 38

Text #16

Max Range: 3.4062. Min Range: -3.4062

Max Act: 3.2020. Min Act: -0.1700

Data Index: 160839 (C4 (Web Text))

Max Activating Token Index: 555

Text #17

Max Range: 3.4062. Min Range: -3.4062

Max Act: 3.1863. Min Act: -0.1700

Data Index: 1402756 (Python Code)

Max Activating Token Index: 135

Text #18

Max Range: 3.4062. Min Range: -3.4062

Max Act: 3.1863. Min Act: -0.1700

Data Index: 1381574 (Python Code)

Max Activating Token Index: 937

Text #19

Max Range: 3.4062. Min Range: -3.4062

Max Act: 3.1863. Min Act: -0.1700

Data Index: 1529036 (Python Code)

Max Activating Token Index: 51