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

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

Neuron 316 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 4.0985. Min Range: -4.0985

Max Act: 4.0985. Min Act: -0.1700

Data Index: 460020 (C4 (Web Text))

Max Activating Token Index: 431

Text #1

Max Range: 4.0985. Min Range: -4.0985

Max Act: 3.9422. Min Act: -0.1700

Data Index: 155077 (C4 (Web Text))

Max Activating Token Index: 322

Text #2

Max Range: 4.0985. Min Range: -4.0985

Max Act: 3.7233. Min Act: -0.1700

Data Index: 1444397 (Python Code)

Max Activating Token Index: 231

Text #3

Max Range: 4.0985. Min Range: -4.0985

Max Act: 3.6763. Min Act: -0.1698

Data Index: 955439 (C4 (Web Text))

Max Activating Token Index: 93

Text #4

Max Range: 4.0985. Min Range: -4.0985

Max Act: 3.6763. Min Act: -0.1700

Data Index: 580100 (C4 (Web Text))

Max Activating Token Index: 206

Text #5

Max Range: 4.0985. Min Range: -4.0985

Max Act: 3.6294. Min Act: -0.1700

Data Index: 1377529 (Python Code)

Max Activating Token Index: 332

Text #6

Max Range: 4.0985. Min Range: -4.0985

Max Act: 3.6294. Min Act: -0.1700

Data Index: 1103120 (C4 (Web Text))

Max Activating Token Index: 561

Text #7

Max Range: 4.0985. Min Range: -4.0985

Max Act: 3.6607. Min Act: -0.1700

Data Index: 637767 (C4 (Web Text))

Max Activating Token Index: 379

Text #8

Max Range: 4.0985. Min Range: -4.0985

Max Act: 3.6137. Min Act: -0.1700

Data Index: 26556 (C4 (Web Text))

Max Activating Token Index: 381

Text #9

Max Range: 4.0985. Min Range: -4.0985

Max Act: 3.5667. Min Act: -0.1700

Data Index: 518584 (C4 (Web Text))

Max Activating Token Index: 198

Text #10

Max Range: 4.0985. Min Range: -4.0985

Max Act: 3.5667. Min Act: -0.1700

Data Index: 1095508 (C4 (Web Text))

Max Activating Token Index: 374

Text #11

Max Range: 4.0985. Min Range: -4.0985

Max Act: 3.5511. Min Act: -0.1699

Data Index: 1081445 (C4 (Web Text))

Max Activating Token Index: 815

Text #12

Max Range: 4.0985. Min Range: -4.0985

Max Act: 3.5511. Min Act: -0.1700

Data Index: 531194 (C4 (Web Text))

Max Activating Token Index: 161

Text #13

Max Range: 4.0985. Min Range: -4.0985

Max Act: 3.5354. Min Act: -0.1700

Data Index: 581470 (C4 (Web Text))

Max Activating Token Index: 798

Text #14

Max Range: 4.0985. Min Range: -4.0985

Max Act: 3.5041. Min Act: -0.1700

Data Index: 819309 (C4 (Web Text))

Max Activating Token Index: 729

Text #15

Max Range: 4.0985. Min Range: -4.0985

Max Act: 3.5511. Min Act: -0.1700

Data Index: 1271387 (C4 (Web Text))

Max Activating Token Index: 845

Text #16

Max Range: 4.0985. Min Range: -4.0985

Max Act: 3.5041. Min Act: -0.1700

Data Index: 566488 (C4 (Web Text))

Max Activating Token Index: 386

Text #17

Max Range: 4.0985. Min Range: -4.0985

Max Act: 3.5041. Min Act: -0.1700

Data Index: 640849 (C4 (Web Text))

Max Activating Token Index: 430

Text #18

Max Range: 4.0985. Min Range: -4.0985

Max Act: 3.4884. Min Act: -0.1700

Data Index: 920956 (C4 (Web Text))

Max Activating Token Index: 849

Text #19

Max Range: 4.0985. Min Range: -4.0985

Max Act: 3.4571. Min Act: -0.1700

Data Index: 578931 (C4 (Web Text))

Max Activating Token Index: 485