Model: GELU Model: 1 Layers, 2048 Neurons per Layer

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

Neuron 1277 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Hooked Transformer Loading: HookedTransformer.from_pretrained('gelu-1l')

Click the toggle to see the full text!

Text #0

Max Range: 4.0665. Min Range: -4.0665

Max Act: 4.0665. Min Act: -0.1687

Data Index: 1274143 (C4 (Web Text))

Max Activating Token Index: 92

Text #1

Max Range: 4.0665. Min Range: -4.0665

Max Act: 4.0329. Min Act: -0.1700

Data Index: 1597140 (Python Code)

Max Activating Token Index: 894

Text #2

Max Range: 4.0665. Min Range: -4.0665

Max Act: 3.9565. Min Act: -0.1699

Data Index: 605475 (C4 (Web Text))

Max Activating Token Index: 695

Text #3

Max Range: 4.0665. Min Range: -4.0665

Max Act: 3.8749. Min Act: -0.1700

Data Index: 1481592 (Python Code)

Max Activating Token Index: 838

Text #4

Max Range: 4.0665. Min Range: -4.0665

Max Act: 3.8747. Min Act: -0.1700

Data Index: 1386598 (Python Code)

Max Activating Token Index: 108

Text #5

Max Range: 4.0665. Min Range: -4.0665

Max Act: 3.8207. Min Act: -0.1700

Data Index: 49313 (C4 (Web Text))

Max Activating Token Index: 117

Text #6

Max Range: 4.0665. Min Range: -4.0665

Max Act: 3.8453. Min Act: -0.1694

Data Index: 240970 (C4 (Web Text))

Max Activating Token Index: 993

Text #7

Max Range: 4.0665. Min Range: -4.0665

Max Act: 3.7918. Min Act: -0.1700

Data Index: 737005 (C4 (Web Text))

Max Activating Token Index: 287

Text #8

Max Range: 4.0665. Min Range: -4.0665

Max Act: 3.8051. Min Act: -0.1700

Data Index: 1560319 (Python Code)

Max Activating Token Index: 200

Text #9

Max Range: 4.0665. Min Range: -4.0665

Max Act: 3.7904. Min Act: -0.1510

Data Index: 443295 (C4 (Web Text))

Max Activating Token Index: 533

Text #10

Max Range: 4.0665. Min Range: -4.0665

Max Act: 3.8101. Min Act: -0.1688

Data Index: 1286780 (C4 (Web Text))

Max Activating Token Index: 373

Text #11

Max Range: 4.0665. Min Range: -4.0665

Max Act: 3.8119. Min Act: -0.1698

Data Index: 230297 (C4 (Web Text))

Max Activating Token Index: 582

Text #12

Max Range: 4.0665. Min Range: -4.0665

Max Act: 3.7620. Min Act: -0.1688

Data Index: 940326 (C4 (Web Text))

Max Activating Token Index: 116

Text #13

Max Range: 4.0665. Min Range: -4.0665

Max Act: 3.7660. Min Act: -0.1510

Data Index: 484261 (C4 (Web Text))

Max Activating Token Index: 512

Text #14

Max Range: 4.0665. Min Range: -4.0665

Max Act: 3.7768. Min Act: -0.1700

Data Index: 941409 (C4 (Web Text))

Max Activating Token Index: 173

Text #15

Max Range: 4.0665. Min Range: -4.0665

Max Act: 3.7587. Min Act: -0.1516

Data Index: 1338870 (C4 (Web Text))

Max Activating Token Index: 846

Text #16

Max Range: 4.0665. Min Range: -4.0665

Max Act: 3.7449. Min Act: -0.1682

Data Index: 395817 (C4 (Web Text))

Max Activating Token Index: 83

Text #17

Max Range: 4.0665. Min Range: -4.0665

Max Act: 3.7289. Min Act: -0.1590

Data Index: 739602 (C4 (Web Text))

Max Activating Token Index: 759

Text #18

Max Range: 4.0665. Min Range: -4.0665

Max Act: 3.7340. Min Act: -0.1698

Data Index: 722789 (C4 (Web Text))

Max Activating Token Index: 318

Text #19

Max Range: 4.0665. Min Range: -4.0665

Max Act: 3.7051. Min Act: -0.1694

Data Index: 728983 (C4 (Web Text))

Max Activating Token Index: 248