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

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

Neuron 1150 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: 5.6633. Min Range: -5.6633

Max Act: 5.6633. Min Act: -0.1700

Data Index: 84404 (C4 (Web Text))

Max Activating Token Index: 173

Text #1

Max Range: 5.6633. Min Range: -5.6633

Max Act: 5.4758. Min Act: -0.1700

Data Index: 722376 (C4 (Web Text))

Max Activating Token Index: 911

Text #2

Max Range: 5.6633. Min Range: -5.6633

Max Act: 5.4758. Min Act: -0.1700

Data Index: 1207872 (C4 (Web Text))

Max Activating Token Index: 493

Text #3

Max Range: 5.6633. Min Range: -5.6633

Max Act: 5.3196. Min Act: -0.1700

Data Index: 454271 (C4 (Web Text))

Max Activating Token Index: 375

Text #4

Max Range: 5.6633. Min Range: -5.6633

Max Act: 5.4446. Min Act: -0.1700

Data Index: 208373 (C4 (Web Text))

Max Activating Token Index: 380

Text #5

Max Range: 5.6633. Min Range: -5.6633

Max Act: 5.3508. Min Act: -0.1700

Data Index: 1240041 (C4 (Web Text))

Max Activating Token Index: 346

Text #6

Max Range: 5.6633. Min Range: -5.6633

Max Act: 5.3508. Min Act: -0.1700

Data Index: 1046312 (C4 (Web Text))

Max Activating Token Index: 683

Text #7

Max Range: 5.6633. Min Range: -5.6633

Max Act: 5.3508. Min Act: -0.1700

Data Index: 496883 (C4 (Web Text))

Max Activating Token Index: 718

Text #8

Max Range: 5.6633. Min Range: -5.6633

Max Act: 5.3508. Min Act: -0.1700

Data Index: 1168920 (C4 (Web Text))

Max Activating Token Index: 239

Text #9

Max Range: 5.6633. Min Range: -5.6633

Max Act: 5.3196. Min Act: -0.1700

Data Index: 900072 (C4 (Web Text))

Max Activating Token Index: 989

Text #10

Max Range: 5.6633. Min Range: -5.6633

Max Act: 5.2571. Min Act: -0.1700

Data Index: 376888 (C4 (Web Text))

Max Activating Token Index: 538

Text #11

Max Range: 5.6633. Min Range: -5.6633

Max Act: 5.2571. Min Act: -0.1700

Data Index: 679673 (C4 (Web Text))

Max Activating Token Index: 595

Text #12

Max Range: 5.6633. Min Range: -5.6633

Max Act: 5.2883. Min Act: -0.1700

Data Index: 662976 (C4 (Web Text))

Max Activating Token Index: 509

Text #13

Max Range: 5.6633. Min Range: -5.6633

Max Act: 5.2571. Min Act: -0.1700

Data Index: 1041385 (C4 (Web Text))

Max Activating Token Index: 265

Text #14

Max Range: 5.6633. Min Range: -5.6633

Max Act: 5.2571. Min Act: -0.1700

Data Index: 582940 (C4 (Web Text))

Max Activating Token Index: 618

Text #15

Max Range: 5.6633. Min Range: -5.6633

Max Act: 5.2258. Min Act: -0.1700

Data Index: 336976 (C4 (Web Text))

Max Activating Token Index: 773

Text #16

Max Range: 5.6633. Min Range: -5.6633

Max Act: 5.0696. Min Act: -0.1700

Data Index: 945164 (C4 (Web Text))

Max Activating Token Index: 862

Text #17

Max Range: 5.6633. Min Range: -5.6633

Max Act: 5.2883. Min Act: -0.1700

Data Index: 116925 (C4 (Web Text))

Max Activating Token Index: 380

Text #18

Max Range: 5.6633. Min Range: -5.6633

Max Act: 5.2258. Min Act: -0.1700

Data Index: 1218413 (C4 (Web Text))

Max Activating Token Index: 126

Text #19

Max Range: 5.6633. Min Range: -5.6633

Max Act: 5.1946. Min Act: -0.1700

Data Index: 802766 (C4 (Web Text))

Max Activating Token Index: 734