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

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

Neuron 521 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 7.2169. Min Range: -7.2169

Max Act: 7.2169. Min Act: -0.1700

Data Index: 1309258 (C4 (Web Text))

Max Activating Token Index: 444

Text #1

Max Range: 7.2169. Min Range: -7.2169

Max Act: 7.0919. Min Act: -0.1700

Data Index: 926497 (C4 (Web Text))

Max Activating Token Index: 76

Text #2

Max Range: 7.2169. Min Range: -7.2169

Max Act: 6.9044. Min Act: -0.1699

Data Index: 213961 (C4 (Web Text))

Max Activating Token Index: 439

Text #3

Max Range: 7.2169. Min Range: -7.2169

Max Act: 6.8419. Min Act: -0.1700

Data Index: 503304 (C4 (Web Text))

Max Activating Token Index: 527

Text #4

Max Range: 7.2169. Min Range: -7.2169

Max Act: 6.9044. Min Act: -0.1699

Data Index: 134177 (C4 (Web Text))

Max Activating Token Index: 398

Text #5

Max Range: 7.2169. Min Range: -7.2169

Max Act: 6.8419. Min Act: -0.1700

Data Index: 1250480 (C4 (Web Text))

Max Activating Token Index: 725

Text #6

Max Range: 7.2169. Min Range: -7.2169

Max Act: 6.9044. Min Act: -0.1699

Data Index: 1133185 (C4 (Web Text))

Max Activating Token Index: 139

Text #7

Max Range: 7.2169. Min Range: -7.2169

Max Act: 6.7794. Min Act: -0.1700

Data Index: 1045832 (C4 (Web Text))

Max Activating Token Index: 169

Text #8

Max Range: 7.2169. Min Range: -7.2169

Max Act: 6.7169. Min Act: -0.1700

Data Index: 1214246 (C4 (Web Text))

Max Activating Token Index: 380

Text #9

Max Range: 7.2169. Min Range: -7.2169

Max Act: 6.7794. Min Act: -0.1700

Data Index: 312684 (C4 (Web Text))

Max Activating Token Index: 315

Text #10

Max Range: 7.2169. Min Range: -7.2169

Max Act: 6.6544. Min Act: -0.1700

Data Index: 674755 (C4 (Web Text))

Max Activating Token Index: 705

Text #11

Max Range: 7.2169. Min Range: -7.2169

Max Act: 6.6544. Min Act: -0.1699

Data Index: 134824 (C4 (Web Text))

Max Activating Token Index: 782

Text #12

Max Range: 7.2169. Min Range: -7.2169

Max Act: 6.7169. Min Act: -0.1700

Data Index: 900297 (C4 (Web Text))

Max Activating Token Index: 970

Text #13

Max Range: 7.2169. Min Range: -7.2169

Max Act: 6.6232. Min Act: -0.1699

Data Index: 41642 (C4 (Web Text))

Max Activating Token Index: 565

Text #14

Max Range: 7.2169. Min Range: -7.2169

Max Act: 6.6544. Min Act: -0.1700

Data Index: 883734 (C4 (Web Text))

Max Activating Token Index: 13

Text #15

Max Range: 7.2169. Min Range: -7.2169

Max Act: 6.6232. Min Act: -0.1699

Data Index: 1184509 (C4 (Web Text))

Max Activating Token Index: 397

Text #16

Max Range: 7.2169. Min Range: -7.2169

Max Act: 6.6857. Min Act: -0.1700

Data Index: 375120 (C4 (Web Text))

Max Activating Token Index: 473

Text #17

Max Range: 7.2169. Min Range: -7.2169

Max Act: 6.5607. Min Act: -0.1700

Data Index: 242115 (C4 (Web Text))

Max Activating Token Index: 315

Text #18

Max Range: 7.2169. Min Range: -7.2169

Max Act: 6.6544. Min Act: -0.1700

Data Index: 1076867 (C4 (Web Text))

Max Activating Token Index: 470

Text #19

Max Range: 7.2169. Min Range: -7.2169

Max Act: 6.6544. Min Act: -0.1700

Data Index: 1123910 (C4 (Web Text))

Max Activating Token Index: 763