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

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

Neuron 393 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: 3.3258. Min Range: -3.3258

Max Act: 3.3258. Min Act: -0.1700

Data Index: 1360152 (Python Code)

Max Activating Token Index: 713

Text #1

Max Range: 3.3258. Min Range: -3.3258

Max Act: 3.2786. Min Act: -0.1700

Data Index: 1059643 (C4 (Web Text))

Max Activating Token Index: 392

Text #2

Max Range: 3.3258. Min Range: -3.3258

Max Act: 3.3101. Min Act: -0.1700

Data Index: 1621362 (Python Code)

Max Activating Token Index: 773

Text #3

Max Range: 3.3258. Min Range: -3.3258

Max Act: 3.2944. Min Act: -0.1700

Data Index: 1623407 (Python Code)

Max Activating Token Index: 931

Text #4

Max Range: 3.3258. Min Range: -3.3258

Max Act: 3.2629. Min Act: -0.1700

Data Index: 1550086 (Python Code)

Max Activating Token Index: 93

Text #5

Max Range: 3.3258. Min Range: -3.3258

Max Act: 3.2315. Min Act: -0.1700

Data Index: 1360773 (Python Code)

Max Activating Token Index: 386

Text #6

Max Range: 3.3258. Min Range: -3.3258

Max Act: 3.2158. Min Act: -0.1700

Data Index: 489277 (C4 (Web Text))

Max Activating Token Index: 808

Text #7

Max Range: 3.3258. Min Range: -3.3258

Max Act: 3.2158. Min Act: -0.1700

Data Index: 1593366 (Python Code)

Max Activating Token Index: 14

Text #8

Max Range: 3.3258. Min Range: -3.3258

Max Act: 3.2158. Min Act: -0.1700

Data Index: 1191195 (C4 (Web Text))

Max Activating Token Index: 188

Text #9

Max Range: 3.3258. Min Range: -3.3258

Max Act: 3.2000. Min Act: -0.1700

Data Index: 1541461 (Python Code)

Max Activating Token Index: 717

Text #10

Max Range: 3.3258. Min Range: -3.3258

Max Act: 3.2158. Min Act: -0.1700

Data Index: 1440099 (Python Code)

Max Activating Token Index: 918

Text #11

Max Range: 3.3258. Min Range: -3.3258

Max Act: 3.1843. Min Act: -0.1700

Data Index: 1533769 (Python Code)

Max Activating Token Index: 317

Text #12

Max Range: 3.3258. Min Range: -3.3258

Max Act: 3.1686. Min Act: -0.1700

Data Index: 1455937 (Python Code)

Max Activating Token Index: 276

Text #13

Max Range: 3.3258. Min Range: -3.3258

Max Act: 3.1843. Min Act: -0.1700

Data Index: 1433215 (Python Code)

Max Activating Token Index: 995

Text #14

Max Range: 3.3258. Min Range: -3.3258

Max Act: 3.1686. Min Act: -0.1700

Data Index: 546391 (C4 (Web Text))

Max Activating Token Index: 303

Text #15

Max Range: 3.3258. Min Range: -3.3258

Max Act: 3.1528. Min Act: -0.1700

Data Index: 1577078 (Python Code)

Max Activating Token Index: 66

Text #16

Max Range: 3.3258. Min Range: -3.3258

Max Act: 3.1686. Min Act: -0.1700

Data Index: 1442331 (Python Code)

Max Activating Token Index: 378

Text #17

Max Range: 3.3258. Min Range: -3.3258

Max Act: 3.1371. Min Act: -0.1700

Data Index: 1366108 (Python Code)

Max Activating Token Index: 392

Text #18

Max Range: 3.3258. Min Range: -3.3258

Max Act: 3.1528. Min Act: -0.1700

Data Index: 1538160 (Python Code)

Max Activating Token Index: 919

Text #19

Max Range: 3.3258. Min Range: -3.3258

Max Act: 3.1371. Min Act: -0.1700

Data Index: 1539036 (Python Code)

Max Activating Token Index: 276