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

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

Neuron 537 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: 5.9672. Min Range: -5.9672

Max Act: 5.9672. Min Act: -0.1700

Data Index: 749290 (C4 (Web Text))

Max Activating Token Index: 751

Text #1

Max Range: 5.9672. Min Range: -5.9672

Max Act: 5.5922. Min Act: -0.1699

Data Index: 1111867 (C4 (Web Text))

Max Activating Token Index: 239

Text #2

Max Range: 5.9672. Min Range: -5.9672

Max Act: 5.5297. Min Act: -0.1700

Data Index: 111463 (C4 (Web Text))

Max Activating Token Index: 309

Text #3

Max Range: 5.9672. Min Range: -5.9672

Max Act: 5.4985. Min Act: -0.1330

Data Index: 312738 (C4 (Web Text))

Max Activating Token Index: 536

Text #4

Max Range: 5.9672. Min Range: -5.9672

Max Act: 5.4985. Min Act: -0.1663

Data Index: 533741 (C4 (Web Text))

Max Activating Token Index: 977

Text #5

Max Range: 5.9672. Min Range: -5.9672

Max Act: 5.4360. Min Act: -0.1676

Data Index: 1264176 (C4 (Web Text))

Max Activating Token Index: 666

Text #6

Max Range: 5.9672. Min Range: -5.9672

Max Act: 5.4360. Min Act: -0.1640

Data Index: 1187418 (C4 (Web Text))

Max Activating Token Index: 173

Text #7

Max Range: 5.9672. Min Range: -5.9672

Max Act: 5.3422. Min Act: -0.1699

Data Index: 1261451 (C4 (Web Text))

Max Activating Token Index: 527

Text #8

Max Range: 5.9672. Min Range: -5.9672

Max Act: 5.3110. Min Act: -0.1482

Data Index: 576182 (C4 (Web Text))

Max Activating Token Index: 262

Text #9

Max Range: 5.9672. Min Range: -5.9672

Max Act: 5.2485. Min Act: -0.1700

Data Index: 1127126 (C4 (Web Text))

Max Activating Token Index: 956

Text #10

Max Range: 5.9672. Min Range: -5.9672

Max Act: 5.2485. Min Act: -0.1700

Data Index: 46958 (C4 (Web Text))

Max Activating Token Index: 680

Text #11

Max Range: 5.9672. Min Range: -5.9672

Max Act: 5.2485. Min Act: -0.1699

Data Index: 682139 (C4 (Web Text))

Max Activating Token Index: 915

Text #12

Max Range: 5.9672. Min Range: -5.9672

Max Act: 5.2172. Min Act: -0.1698

Data Index: 822692 (C4 (Web Text))

Max Activating Token Index: 910

Text #13

Max Range: 5.9672. Min Range: -5.9672

Max Act: 5.1547. Min Act: -0.1700

Data Index: 1188184 (C4 (Web Text))

Max Activating Token Index: 18

Text #14

Max Range: 5.9672. Min Range: -5.9672

Max Act: 5.1860. Min Act: -0.1700

Data Index: 168653 (C4 (Web Text))

Max Activating Token Index: 918

Text #15

Max Range: 5.9672. Min Range: -5.9672

Max Act: 5.2172. Min Act: -0.1699

Data Index: 635140 (C4 (Web Text))

Max Activating Token Index: 362

Text #16

Max Range: 5.9672. Min Range: -5.9672

Max Act: 5.1860. Min Act: -0.1700

Data Index: 1222700 (C4 (Web Text))

Max Activating Token Index: 661

Text #17

Max Range: 5.9672. Min Range: -5.9672

Max Act: 5.1547. Min Act: -0.1700

Data Index: 495050 (C4 (Web Text))

Max Activating Token Index: 408

Text #18

Max Range: 5.9672. Min Range: -5.9672

Max Act: 5.1547. Min Act: -0.1700

Data Index: 1146006 (C4 (Web Text))

Max Activating Token Index: 121

Text #19

Max Range: 5.9672. Min Range: -5.9672

Max Act: 5.1547. Min Act: -0.1698

Data Index: 628281 (C4 (Web Text))

Max Activating Token Index: 695