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

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

Neuron 2024 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 7.9663. Min Range: -7.9663

Max Act: 7.9663. Min Act: -0.1700

Data Index: 724602 (C4 (Web Text))

Max Activating Token Index: 490

Text #1

Max Range: 7.9663. Min Range: -7.9663

Max Act: 7.7163. Min Act: -0.1700

Data Index: 1321435 (C4 (Web Text))

Max Activating Token Index: 363

Text #2

Max Range: 7.9663. Min Range: -7.9663

Max Act: 7.6225. Min Act: -0.1700

Data Index: 1335679 (C4 (Web Text))

Max Activating Token Index: 711

Text #3

Max Range: 7.9663. Min Range: -7.9663

Max Act: 7.5288. Min Act: -0.1700

Data Index: 911764 (C4 (Web Text))

Max Activating Token Index: 443

Text #4

Max Range: 7.9663. Min Range: -7.9663

Max Act: 7.6225. Min Act: -0.1700

Data Index: 50596 (C4 (Web Text))

Max Activating Token Index: 556

Text #5

Max Range: 7.9663. Min Range: -7.9663

Max Act: 7.5913. Min Act: -0.1700

Data Index: 1260580 (C4 (Web Text))

Max Activating Token Index: 342

Text #6

Max Range: 7.9663. Min Range: -7.9663

Max Act: 7.5600. Min Act: -0.1700

Data Index: 1124457 (C4 (Web Text))

Max Activating Token Index: 315

Text #7

Max Range: 7.9663. Min Range: -7.9663

Max Act: 7.4975. Min Act: -0.1700

Data Index: 1063620 (C4 (Web Text))

Max Activating Token Index: 534

Text #8

Max Range: 7.9663. Min Range: -7.9663

Max Act: 7.5600. Min Act: -0.1700

Data Index: 229217 (C4 (Web Text))

Max Activating Token Index: 447

Text #9

Max Range: 7.9663. Min Range: -7.9663

Max Act: 7.5288. Min Act: -0.1700

Data Index: 412190 (C4 (Web Text))

Max Activating Token Index: 430

Text #10

Max Range: 7.9663. Min Range: -7.9663

Max Act: 7.5288. Min Act: -0.1700

Data Index: 418868 (C4 (Web Text))

Max Activating Token Index: 534

Text #11

Max Range: 7.9663. Min Range: -7.9663

Max Act: 7.5600. Min Act: -0.1700

Data Index: 934395 (C4 (Web Text))

Max Activating Token Index: 430

Text #12

Max Range: 7.9663. Min Range: -7.9663

Max Act: 7.5288. Min Act: -0.1700

Data Index: 1357685 (C4 (Web Text))

Max Activating Token Index: 686

Text #13

Max Range: 7.9663. Min Range: -7.9663

Max Act: 7.5288. Min Act: -0.1700

Data Index: 719805 (C4 (Web Text))

Max Activating Token Index: 580

Text #14

Max Range: 7.9663. Min Range: -7.9663

Max Act: 7.5600. Min Act: -0.1700

Data Index: 440129 (C4 (Web Text))

Max Activating Token Index: 292

Text #15

Max Range: 7.9663. Min Range: -7.9663

Max Act: 7.4350. Min Act: -0.1700

Data Index: 806887 (C4 (Web Text))

Max Activating Token Index: 323

Text #16

Max Range: 7.9663. Min Range: -7.9663

Max Act: 7.4663. Min Act: -0.1700

Data Index: 1305095 (C4 (Web Text))

Max Activating Token Index: 883

Text #17

Max Range: 7.9663. Min Range: -7.9663

Max Act: 7.4038. Min Act: -0.1700

Data Index: 1323756 (C4 (Web Text))

Max Activating Token Index: 461

Text #18

Max Range: 7.9663. Min Range: -7.9663

Max Act: 7.4663. Min Act: -0.1700

Data Index: 388053 (C4 (Web Text))

Max Activating Token Index: 309

Text #19

Max Range: 7.9663. Min Range: -7.9663

Max Act: 7.5288. Min Act: -0.1700

Data Index: 556288 (C4 (Web Text))

Max Activating Token Index: 298