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

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

Neuron 538 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: 5.4033. Min Range: -5.4033

Max Act: 5.4033. Min Act: -0.1700

Data Index: 600955 (C4 (Web Text))

Max Activating Token Index: 740

Text #1

Max Range: 5.4033. Min Range: -5.4033

Max Act: 5.4033. Min Act: -0.1700

Data Index: 471007 (C4 (Web Text))

Max Activating Token Index: 148

Text #2

Max Range: 5.4033. Min Range: -5.4033

Max Act: 5.4033. Min Act: -0.1700

Data Index: 548530 (C4 (Web Text))

Max Activating Token Index: 390

Text #3

Max Range: 5.4033. Min Range: -5.4033

Max Act: 5.3408. Min Act: -0.1700

Data Index: 111907 (C4 (Web Text))

Max Activating Token Index: 371

Text #4

Max Range: 5.4033. Min Range: -5.4033

Max Act: 5.2783. Min Act: -0.1700

Data Index: 724123 (C4 (Web Text))

Max Activating Token Index: 366

Text #5

Max Range: 5.4033. Min Range: -5.4033

Max Act: 5.3720. Min Act: -0.1700

Data Index: 1155034 (C4 (Web Text))

Max Activating Token Index: 803

Text #6

Max Range: 5.4033. Min Range: -5.4033

Max Act: 5.1533. Min Act: -0.1700

Data Index: 470910 (C4 (Web Text))

Max Activating Token Index: 283

Text #7

Max Range: 5.4033. Min Range: -5.4033

Max Act: 5.1533. Min Act: -0.1700

Data Index: 310839 (C4 (Web Text))

Max Activating Token Index: 908

Text #8

Max Range: 5.4033. Min Range: -5.4033

Max Act: 5.1220. Min Act: -0.1700

Data Index: 964894 (C4 (Web Text))

Max Activating Token Index: 207

Text #9

Max Range: 5.4033. Min Range: -5.4033

Max Act: 5.1220. Min Act: -0.1700

Data Index: 929740 (C4 (Web Text))

Max Activating Token Index: 546

Text #10

Max Range: 5.4033. Min Range: -5.4033

Max Act: 5.0908. Min Act: -0.1700

Data Index: 607283 (C4 (Web Text))

Max Activating Token Index: 857

Text #11

Max Range: 5.4033. Min Range: -5.4033

Max Act: 5.1220. Min Act: -0.1700

Data Index: 997189 (C4 (Web Text))

Max Activating Token Index: 996

Text #12

Max Range: 5.4033. Min Range: -5.4033

Max Act: 5.0595. Min Act: -0.1700

Data Index: 76969 (C4 (Web Text))

Max Activating Token Index: 670

Text #13

Max Range: 5.4033. Min Range: -5.4033

Max Act: 5.0595. Min Act: -0.1700

Data Index: 894604 (C4 (Web Text))

Max Activating Token Index: 433

Text #14

Max Range: 5.4033. Min Range: -5.4033

Max Act: 5.0908. Min Act: -0.1700

Data Index: 180071 (C4 (Web Text))

Max Activating Token Index: 811

Text #15

Max Range: 5.4033. Min Range: -5.4033

Max Act: 5.0595. Min Act: -0.1700

Data Index: 142300 (C4 (Web Text))

Max Activating Token Index: 81

Text #16

Max Range: 5.4033. Min Range: -5.4033

Max Act: 5.1220. Min Act: -0.1700

Data Index: 1187584 (C4 (Web Text))

Max Activating Token Index: 814

Text #17

Max Range: 5.4033. Min Range: -5.4033

Max Act: 5.0595. Min Act: -0.1700

Data Index: 1078338 (C4 (Web Text))

Max Activating Token Index: 229

Text #18

Max Range: 5.4033. Min Range: -5.4033

Max Act: 5.0283. Min Act: -0.1700

Data Index: 1182237 (C4 (Web Text))

Max Activating Token Index: 473

Text #19

Max Range: 5.4033. Min Range: -5.4033

Max Act: 5.0283. Min Act: -0.1700

Data Index: 299545 (C4 (Web Text))

Max Activating Token Index: 559