Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 2020 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 5.6478. Min Range: -5.6478

Max Act: 5.6478. Min Act: -0.1700

Data Index: 1180560 (C4 (Web Text))

Max Activating Token Index: 966

Text #1

Max Range: 5.6478. Min Range: -5.6478

Max Act: 5.5541. Min Act: -0.1700

Data Index: 1045235 (C4 (Web Text))

Max Activating Token Index: 768

Text #2

Max Range: 5.6478. Min Range: -5.6478

Max Act: 5.5541. Min Act: -0.1700

Data Index: 956684 (C4 (Web Text))

Max Activating Token Index: 333

Text #3

Max Range: 5.6478. Min Range: -5.6478

Max Act: 5.4916. Min Act: -0.1700

Data Index: 434419 (C4 (Web Text))

Max Activating Token Index: 528

Text #4

Max Range: 5.6478. Min Range: -5.6478

Max Act: 5.4603. Min Act: -0.1700

Data Index: 732214 (C4 (Web Text))

Max Activating Token Index: 643

Text #5

Max Range: 5.6478. Min Range: -5.6478

Max Act: 5.3666. Min Act: -0.1700

Data Index: 576244 (C4 (Web Text))

Max Activating Token Index: 82

Text #6

Max Range: 5.6478. Min Range: -5.6478

Max Act: 5.3666. Min Act: -0.1700

Data Index: 1170091 (C4 (Web Text))

Max Activating Token Index: 617

Text #7

Max Range: 5.6478. Min Range: -5.6478

Max Act: 5.3666. Min Act: -0.1700

Data Index: 1274730 (C4 (Web Text))

Max Activating Token Index: 673

Text #8

Max Range: 5.6478. Min Range: -5.6478

Max Act: 5.3978. Min Act: -0.1700

Data Index: 1012574 (C4 (Web Text))

Max Activating Token Index: 646

Text #9

Max Range: 5.6478. Min Range: -5.6478

Max Act: 5.3666. Min Act: -0.1699

Data Index: 1050521 (C4 (Web Text))

Max Activating Token Index: 302

Text #10

Max Range: 5.6478. Min Range: -5.6478

Max Act: 5.2416. Min Act: -0.1700

Data Index: 503086 (C4 (Web Text))

Max Activating Token Index: 927

Text #11

Max Range: 5.6478. Min Range: -5.6478

Max Act: 5.3041. Min Act: -0.1700

Data Index: 758207 (C4 (Web Text))

Max Activating Token Index: 964

Text #12

Max Range: 5.6478. Min Range: -5.6478

Max Act: 5.2416. Min Act: -0.1700

Data Index: 666635 (C4 (Web Text))

Max Activating Token Index: 524

Text #13

Max Range: 5.6478. Min Range: -5.6478

Max Act: 5.3041. Min Act: -0.1700

Data Index: 1219915 (C4 (Web Text))

Max Activating Token Index: 907

Text #14

Max Range: 5.6478. Min Range: -5.6478

Max Act: 5.2728. Min Act: -0.1700

Data Index: 492667 (C4 (Web Text))

Max Activating Token Index: 64

Text #15

Max Range: 5.6478. Min Range: -5.6478

Max Act: 5.2416. Min Act: -0.1700

Data Index: 231539 (C4 (Web Text))

Max Activating Token Index: 614

Text #16

Max Range: 5.6478. Min Range: -5.6478

Max Act: 5.2416. Min Act: -0.1700

Data Index: 1015746 (C4 (Web Text))

Max Activating Token Index: 398

Text #17

Max Range: 5.6478. Min Range: -5.6478

Max Act: 5.2728. Min Act: -0.1700

Data Index: 439279 (C4 (Web Text))

Max Activating Token Index: 913

Text #18

Max Range: 5.6478. Min Range: -5.6478

Max Act: 5.2103. Min Act: -0.1700

Data Index: 308651 (C4 (Web Text))

Max Activating Token Index: 572

Text #19

Max Range: 5.6478. Min Range: -5.6478

Max Act: 5.2416. Min Act: -0.1699

Data Index: 843521 (C4 (Web Text))

Max Activating Token Index: 451