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

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

Neuron 1283 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.9597. Min Range: -5.9597

Max Act: 5.9597. Min Act: -0.1700

Data Index: 841378 (C4 (Web Text))

Max Activating Token Index: 821

Text #1

Max Range: 5.9597. Min Range: -5.9597

Max Act: 5.8347. Min Act: -0.1700

Data Index: 1220500 (C4 (Web Text))

Max Activating Token Index: 421

Text #2

Max Range: 5.9597. Min Range: -5.9597

Max Act: 5.8347. Min Act: -0.1700

Data Index: 975462 (C4 (Web Text))

Max Activating Token Index: 720

Text #3

Max Range: 5.9597. Min Range: -5.9597

Max Act: 5.7722. Min Act: -0.1700

Data Index: 392979 (C4 (Web Text))

Max Activating Token Index: 648

Text #4

Max Range: 5.9597. Min Range: -5.9597

Max Act: 5.9597. Min Act: -0.1700

Data Index: 1136770 (C4 (Web Text))

Max Activating Token Index: 320

Text #5

Max Range: 5.9597. Min Range: -5.9597

Max Act: 5.7097. Min Act: -0.1700

Data Index: 1115134 (C4 (Web Text))

Max Activating Token Index: 993

Text #6

Max Range: 5.9597. Min Range: -5.9597

Max Act: 5.6472. Min Act: -0.1700

Data Index: 190426 (C4 (Web Text))

Max Activating Token Index: 504

Text #7

Max Range: 5.9597. Min Range: -5.9597

Max Act: 5.7410. Min Act: -0.1700

Data Index: 800479 (C4 (Web Text))

Max Activating Token Index: 972

Text #8

Max Range: 5.9597. Min Range: -5.9597

Max Act: 5.5535. Min Act: -0.1700

Data Index: 1186915 (C4 (Web Text))

Max Activating Token Index: 202

Text #9

Max Range: 5.9597. Min Range: -5.9597

Max Act: 5.5847. Min Act: -0.1700

Data Index: 528474 (C4 (Web Text))

Max Activating Token Index: 624

Text #10

Max Range: 5.9597. Min Range: -5.9597

Max Act: 5.5535. Min Act: -0.1700

Data Index: 62816 (C4 (Web Text))

Max Activating Token Index: 679

Text #11

Max Range: 5.9597. Min Range: -5.9597

Max Act: 5.5222. Min Act: -0.1700

Data Index: 242717 (C4 (Web Text))

Max Activating Token Index: 500

Text #12

Max Range: 5.9597. Min Range: -5.9597

Max Act: 5.6160. Min Act: -0.1700

Data Index: 819440 (C4 (Web Text))

Max Activating Token Index: 135

Text #13

Max Range: 5.9597. Min Range: -5.9597

Max Act: 5.7722. Min Act: -0.1699

Data Index: 1160556 (C4 (Web Text))

Max Activating Token Index: 403

Text #14

Max Range: 5.9597. Min Range: -5.9597

Max Act: 5.5535. Min Act: -0.1697

Data Index: 1353078 (C4 (Web Text))

Max Activating Token Index: 275

Text #15

Max Range: 5.9597. Min Range: -5.9597

Max Act: 5.3347. Min Act: -0.1700

Data Index: 727863 (C4 (Web Text))

Max Activating Token Index: 415

Text #16

Max Range: 5.9597. Min Range: -5.9597

Max Act: 5.7722. Min Act: -0.1700

Data Index: 302952 (C4 (Web Text))

Max Activating Token Index: 727

Text #17

Max Range: 5.9597. Min Range: -5.9597

Max Act: 5.5847. Min Act: -0.1700

Data Index: 1347289 (C4 (Web Text))

Max Activating Token Index: 1006

Text #18

Max Range: 5.9597. Min Range: -5.9597

Max Act: 5.6160. Min Act: -0.1700

Data Index: 194034 (C4 (Web Text))

Max Activating Token Index: 651

Text #19

Max Range: 5.9597. Min Range: -5.9597

Max Act: 5.4910. Min Act: -0.1699

Data Index: 1561916 (Python Code)

Max Activating Token Index: 887