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

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

Neuron 1622 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: 8.3943. Min Range: -8.3943

Max Act: 8.3943. Min Act: -0.1700

Data Index: 258085 (C4 (Web Text))

Max Activating Token Index: 460

Text #1

Max Range: 8.3943. Min Range: -8.3943

Max Act: 8.3318. Min Act: -0.1700

Data Index: 706656 (C4 (Web Text))

Max Activating Token Index: 648

Text #2

Max Range: 8.3943. Min Range: -8.3943

Max Act: 8.2381. Min Act: -0.1700

Data Index: 910008 (C4 (Web Text))

Max Activating Token Index: 527

Text #3

Max Range: 8.3943. Min Range: -8.3943

Max Act: 8.1756. Min Act: -0.1700

Data Index: 445436 (C4 (Web Text))

Max Activating Token Index: 861

Text #4

Max Range: 8.3943. Min Range: -8.3943

Max Act: 7.9881. Min Act: -0.1700

Data Index: 434585 (C4 (Web Text))

Max Activating Token Index: 764

Text #5

Max Range: 8.3943. Min Range: -8.3943

Max Act: 8.0193. Min Act: -0.1700

Data Index: 1260692 (C4 (Web Text))

Max Activating Token Index: 473

Text #6

Max Range: 8.3943. Min Range: -8.3943

Max Act: 7.8318. Min Act: -0.1700

Data Index: 1011324 (C4 (Web Text))

Max Activating Token Index: 404

Text #7

Max Range: 8.3943. Min Range: -8.3943

Max Act: 7.7381. Min Act: -0.1700

Data Index: 1240335 (C4 (Web Text))

Max Activating Token Index: 584

Text #8

Max Range: 8.3943. Min Range: -8.3943

Max Act: 7.7068. Min Act: -0.1700

Data Index: 740059 (C4 (Web Text))

Max Activating Token Index: 220

Text #9

Max Range: 8.3943. Min Range: -8.3943

Max Act: 7.7381. Min Act: -0.1700

Data Index: 1223233 (C4 (Web Text))

Max Activating Token Index: 157

Text #10

Max Range: 8.3943. Min Range: -8.3943

Max Act: 7.7068. Min Act: -0.1700

Data Index: 198233 (C4 (Web Text))

Max Activating Token Index: 743

Text #11

Max Range: 8.3943. Min Range: -8.3943

Max Act: 7.6756. Min Act: -0.1700

Data Index: 812182 (C4 (Web Text))

Max Activating Token Index: 707

Text #12

Max Range: 8.3943. Min Range: -8.3943

Max Act: 7.6443. Min Act: -0.1700

Data Index: 666761 (C4 (Web Text))

Max Activating Token Index: 1013

Text #13

Max Range: 8.3943. Min Range: -8.3943

Max Act: 7.6443. Min Act: -0.1700

Data Index: 1133685 (C4 (Web Text))

Max Activating Token Index: 912

Text #14

Max Range: 8.3943. Min Range: -8.3943

Max Act: 7.6131. Min Act: -0.1700

Data Index: 1192262 (C4 (Web Text))

Max Activating Token Index: 459

Text #15

Max Range: 8.3943. Min Range: -8.3943

Max Act: 7.6131. Min Act: -0.1700

Data Index: 425789 (C4 (Web Text))

Max Activating Token Index: 488

Text #16

Max Range: 8.3943. Min Range: -8.3943

Max Act: 7.5818. Min Act: -0.1700

Data Index: 261189 (C4 (Web Text))

Max Activating Token Index: 681

Text #17

Max Range: 8.3943. Min Range: -8.3943

Max Act: 7.5506. Min Act: -0.1700

Data Index: 266331 (C4 (Web Text))

Max Activating Token Index: 504

Text #18

Max Range: 8.3943. Min Range: -8.3943

Max Act: 7.5818. Min Act: -0.1700

Data Index: 719921 (C4 (Web Text))

Max Activating Token Index: 903

Text #19

Max Range: 8.3943. Min Range: -8.3943

Max Act: 7.5193. Min Act: -0.1700

Data Index: 127452 (C4 (Web Text))

Max Activating Token Index: 168