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

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

Neuron 740 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.5336. Min Range: -3.5336

Max Act: 3.5336. Min Act: -0.1700

Data Index: 617064 (C4 (Web Text))

Max Activating Token Index: 404

Text #1

Max Range: 3.5336. Min Range: -3.5336

Max Act: 3.5179. Min Act: -0.1700

Data Index: 1014407 (C4 (Web Text))

Max Activating Token Index: 168

Text #2

Max Range: 3.5336. Min Range: -3.5336

Max Act: 3.5023. Min Act: -0.1700

Data Index: 1598855 (Python Code)

Max Activating Token Index: 163

Text #3

Max Range: 3.5336. Min Range: -3.5336

Max Act: 3.4396. Min Act: -0.1700

Data Index: 1375734 (Python Code)

Max Activating Token Index: 755

Text #4

Max Range: 3.5336. Min Range: -3.5336

Max Act: 3.4553. Min Act: -0.1700

Data Index: 1152409 (C4 (Web Text))

Max Activating Token Index: 452

Text #5

Max Range: 3.5336. Min Range: -3.5336

Max Act: 3.4553. Min Act: -0.1700

Data Index: 21373 (C4 (Web Text))

Max Activating Token Index: 244

Text #6

Max Range: 3.5336. Min Range: -3.5336

Max Act: 3.4396. Min Act: -0.1700

Data Index: 1367657 (Python Code)

Max Activating Token Index: 37

Text #7

Max Range: 3.5336. Min Range: -3.5336

Max Act: 3.4553. Min Act: -0.1700

Data Index: 1067208 (C4 (Web Text))

Max Activating Token Index: 38

Text #8

Max Range: 3.5336. Min Range: -3.5336

Max Act: 3.4239. Min Act: -0.1700

Data Index: 1616868 (Python Code)

Max Activating Token Index: 402

Text #9

Max Range: 3.5336. Min Range: -3.5336

Max Act: 3.4239. Min Act: -0.1700

Data Index: 1620235 (Python Code)

Max Activating Token Index: 6

Text #10

Max Range: 3.5336. Min Range: -3.5336

Max Act: 3.4239. Min Act: -0.1700

Data Index: 233068 (C4 (Web Text))

Max Activating Token Index: 975

Text #11

Max Range: 3.5336. Min Range: -3.5336

Max Act: 3.4082. Min Act: -0.1700

Data Index: 1025720 (C4 (Web Text))

Max Activating Token Index: 583

Text #12

Max Range: 3.5336. Min Range: -3.5336

Max Act: 3.3925. Min Act: -0.1700

Data Index: 78510 (C4 (Web Text))

Max Activating Token Index: 775

Text #13

Max Range: 3.5336. Min Range: -3.5336

Max Act: 3.4082. Min Act: -0.1700

Data Index: 1536409 (Python Code)

Max Activating Token Index: 183

Text #14

Max Range: 3.5336. Min Range: -3.5336

Max Act: 3.3925. Min Act: -0.1700

Data Index: 552355 (C4 (Web Text))

Max Activating Token Index: 962

Text #15

Max Range: 3.5336. Min Range: -3.5336

Max Act: 3.3925. Min Act: -0.1700

Data Index: 1395243 (Python Code)

Max Activating Token Index: 1004

Text #16

Max Range: 3.5336. Min Range: -3.5336

Max Act: 3.3612. Min Act: -0.1700

Data Index: 1396292 (Python Code)

Max Activating Token Index: 945

Text #17

Max Range: 3.5336. Min Range: -3.5336

Max Act: 3.4082. Min Act: -0.1700

Data Index: 1049319 (C4 (Web Text))

Max Activating Token Index: 964

Text #18

Max Range: 3.5336. Min Range: -3.5336

Max Act: 3.4082. Min Act: -0.1700

Data Index: 799093 (C4 (Web Text))

Max Activating Token Index: 485

Text #19

Max Range: 3.5336. Min Range: -3.5336

Max Act: 3.3925. Min Act: -0.1700

Data Index: 935674 (C4 (Web Text))

Max Activating Token Index: 803