Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 976 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 3.3321. Min Range: -3.3321

Max Act: 3.3321. Min Act: -0.1700

Data Index: 1622819 (Python Code)

Max Activating Token Index: 463

Text #1

Max Range: 3.3321. Min Range: -3.3321

Max Act: 3.3007. Min Act: -0.1700

Data Index: 592638 (C4 (Web Text))

Max Activating Token Index: 345

Text #2

Max Range: 3.3321. Min Range: -3.3321

Max Act: 3.2693. Min Act: -0.1700

Data Index: 624308 (C4 (Web Text))

Max Activating Token Index: 822

Text #3

Max Range: 3.3321. Min Range: -3.3321

Max Act: 3.2693. Min Act: -0.1700

Data Index: 180800 (C4 (Web Text))

Max Activating Token Index: 172

Text #4

Max Range: 3.3321. Min Range: -3.3321

Max Act: 3.2379. Min Act: -0.1700

Data Index: 348007 (C4 (Web Text))

Max Activating Token Index: 181

Text #5

Max Range: 3.3321. Min Range: -3.3321

Max Act: 3.2379. Min Act: -0.1700

Data Index: 928124 (C4 (Web Text))

Max Activating Token Index: 710

Text #6

Max Range: 3.3321. Min Range: -3.3321

Max Act: 3.2064. Min Act: -0.1700

Data Index: 415890 (C4 (Web Text))

Max Activating Token Index: 269

Text #7

Max Range: 3.3321. Min Range: -3.3321

Max Act: 3.2064. Min Act: -0.1700

Data Index: 944853 (C4 (Web Text))

Max Activating Token Index: 268

Text #8

Max Range: 3.3321. Min Range: -3.3321

Max Act: 3.2221. Min Act: -0.1700

Data Index: 494898 (C4 (Web Text))

Max Activating Token Index: 927

Text #9

Max Range: 3.3321. Min Range: -3.3321

Max Act: 3.2221. Min Act: -0.1700

Data Index: 1645905 (Python Code)

Max Activating Token Index: 321

Text #10

Max Range: 3.3321. Min Range: -3.3321

Max Act: 3.2064. Min Act: -0.1700

Data Index: 259741 (C4 (Web Text))

Max Activating Token Index: 360

Text #11

Max Range: 3.3321. Min Range: -3.3321

Max Act: 3.1749. Min Act: -0.1700

Data Index: 1216523 (C4 (Web Text))

Max Activating Token Index: 312

Text #12

Max Range: 3.3321. Min Range: -3.3321

Max Act: 3.1907. Min Act: -0.1700

Data Index: 1542112 (Python Code)

Max Activating Token Index: 336

Text #13

Max Range: 3.3321. Min Range: -3.3321

Max Act: 3.1907. Min Act: -0.1700

Data Index: 1163225 (C4 (Web Text))

Max Activating Token Index: 584

Text #14

Max Range: 3.3321. Min Range: -3.3321

Max Act: 3.1434. Min Act: -0.1700

Data Index: 423075 (C4 (Web Text))

Max Activating Token Index: 1016

Text #15

Max Range: 3.3321. Min Range: -3.3321

Max Act: 3.1592. Min Act: -0.1699

Data Index: 823451 (C4 (Web Text))

Max Activating Token Index: 50

Text #16

Max Range: 3.3321. Min Range: -3.3321

Max Act: 3.1434. Min Act: -0.1700

Data Index: 1140900 (C4 (Web Text))

Max Activating Token Index: 309

Text #17

Max Range: 3.3321. Min Range: -3.3321

Max Act: 3.1592. Min Act: -0.1700

Data Index: 61313 (C4 (Web Text))

Max Activating Token Index: 36

Text #18

Max Range: 3.3321. Min Range: -3.3321

Max Act: 3.1434. Min Act: -0.1699

Data Index: 245031 (C4 (Web Text))

Max Activating Token Index: 807

Text #19

Max Range: 3.3321. Min Range: -3.3321

Max Act: 3.1749. Min Act: -0.1700

Data Index: 688205 (C4 (Web Text))

Max Activating Token Index: 750