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

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

Neuron 684 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 3.6933. Min Range: -3.6933

Max Act: 3.6933. Min Act: -0.1700

Data Index: 1036672 (C4 (Web Text))

Max Activating Token Index: 774

Text #1

Max Range: 3.6933. Min Range: -3.6933

Max Act: 3.6464. Min Act: -0.1700

Data Index: 722245 (C4 (Web Text))

Max Activating Token Index: 323

Text #2

Max Range: 3.6933. Min Range: -3.6933

Max Act: 3.6620. Min Act: -0.1700

Data Index: 1059264 (C4 (Web Text))

Max Activating Token Index: 782

Text #3

Max Range: 3.6933. Min Range: -3.6933

Max Act: 3.6620. Min Act: -0.1700

Data Index: 214289 (C4 (Web Text))

Max Activating Token Index: 768

Text #4

Max Range: 3.6933. Min Range: -3.6933

Max Act: 3.6464. Min Act: -0.1700

Data Index: 766980 (C4 (Web Text))

Max Activating Token Index: 918

Text #5

Max Range: 3.6933. Min Range: -3.6933

Max Act: 3.5994. Min Act: -0.1700

Data Index: 389586 (C4 (Web Text))

Max Activating Token Index: 570

Text #6

Max Range: 3.6933. Min Range: -3.6933

Max Act: 3.6151. Min Act: -0.1700

Data Index: 1104536 (C4 (Web Text))

Max Activating Token Index: 767

Text #7

Max Range: 3.6933. Min Range: -3.6933

Max Act: 3.6307. Min Act: -0.1700

Data Index: 1276719 (C4 (Web Text))

Max Activating Token Index: 358

Text #8

Max Range: 3.6933. Min Range: -3.6933

Max Act: 3.5994. Min Act: -0.1700

Data Index: 423057 (C4 (Web Text))

Max Activating Token Index: 960

Text #9

Max Range: 3.6933. Min Range: -3.6933

Max Act: 3.5994. Min Act: -0.1700

Data Index: 1039142 (C4 (Web Text))

Max Activating Token Index: 237

Text #10

Max Range: 3.6933. Min Range: -3.6933

Max Act: 3.5837. Min Act: -0.1700

Data Index: 599407 (C4 (Web Text))

Max Activating Token Index: 596

Text #11

Max Range: 3.6933. Min Range: -3.6933

Max Act: 3.5837. Min Act: -0.1700

Data Index: 1136665 (C4 (Web Text))

Max Activating Token Index: 947

Text #12

Max Range: 3.6933. Min Range: -3.6933

Max Act: 3.6151. Min Act: -0.1700

Data Index: 1143044 (C4 (Web Text))

Max Activating Token Index: 834

Text #13

Max Range: 3.6933. Min Range: -3.6933

Max Act: 3.6151. Min Act: -0.1700

Data Index: 802490 (C4 (Web Text))

Max Activating Token Index: 915

Text #14

Max Range: 3.6933. Min Range: -3.6933

Max Act: 3.5994. Min Act: -0.1700

Data Index: 903030 (C4 (Web Text))

Max Activating Token Index: 327

Text #15

Max Range: 3.6933. Min Range: -3.6933

Max Act: 3.6151. Min Act: -0.1700

Data Index: 1181974 (C4 (Web Text))

Max Activating Token Index: 1003

Text #16

Max Range: 3.6933. Min Range: -3.6933

Max Act: 3.5837. Min Act: -0.1700

Data Index: 1192193 (C4 (Web Text))

Max Activating Token Index: 782

Text #17

Max Range: 3.6933. Min Range: -3.6933

Max Act: 3.5994. Min Act: -0.1700

Data Index: 403518 (C4 (Web Text))

Max Activating Token Index: 928

Text #18

Max Range: 3.6933. Min Range: -3.6933

Max Act: 3.6307. Min Act: -0.1700

Data Index: 706424 (C4 (Web Text))

Max Activating Token Index: 556

Text #19

Max Range: 3.6933. Min Range: -3.6933

Max Act: 3.5994. Min Act: -0.1700

Data Index: 1080508 (C4 (Web Text))

Max Activating Token Index: 950