Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 983 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 5.7686. Min Range: -5.7686

Max Act: 5.7686. Min Act: -0.0001

Data Index: 69841 (C4 (Web Text))

Max Activating Token Index: 831

Text #1

Max Range: 5.7686. Min Range: -5.7686

Max Act: 4.7705. Min Act: -0.0001

Data Index: 256206 (C4 (Web Text))

Max Activating Token Index: 130

Text #2

Max Range: 5.7686. Min Range: -5.7686

Max Act: 4.4770. Min Act: -0.0001

Data Index: 568387 (C4 (Web Text))

Max Activating Token Index: 130

Text #3

Max Range: 5.7686. Min Range: -5.7686

Max Act: 4.4708. Min Act: -0.0001

Data Index: 915777 (C4 (Web Text))

Max Activating Token Index: 54

Text #4

Max Range: 5.7686. Min Range: -5.7686

Max Act: 4.1666. Min Act: -0.0001

Data Index: 131353 (C4 (Web Text))

Max Activating Token Index: 592

Text #5

Max Range: 5.7686. Min Range: -5.7686

Max Act: 4.1500. Min Act: -0.0001

Data Index: 757160 (C4 (Web Text))

Max Activating Token Index: 133

Text #6

Max Range: 5.7686. Min Range: -5.7686

Max Act: 4.0476. Min Act: -0.0001

Data Index: 167804 (C4 (Web Text))

Max Activating Token Index: 690

Text #7

Max Range: 5.7686. Min Range: -5.7686

Max Act: 4.0434. Min Act: -0.0001

Data Index: 185697 (C4 (Web Text))

Max Activating Token Index: 793

Text #8

Max Range: 5.7686. Min Range: -5.7686

Max Act: 3.6778. Min Act: -0.0001

Data Index: 1215897 (C4 (Web Text))

Max Activating Token Index: 116

Text #9

Max Range: 5.7686. Min Range: -5.7686

Max Act: 3.9437. Min Act: -0.0001

Data Index: 988623 (C4 (Web Text))

Max Activating Token Index: 626

Text #10

Max Range: 5.7686. Min Range: -5.7686

Max Act: 3.8774. Min Act: -0.0001

Data Index: 617201 (C4 (Web Text))

Max Activating Token Index: 990

Text #11

Max Range: 5.7686. Min Range: -5.7686

Max Act: 3.5513. Min Act: -0.0001

Data Index: 440690 (C4 (Web Text))

Max Activating Token Index: 673

Text #12

Max Range: 5.7686. Min Range: -5.7686

Max Act: 3.8386. Min Act: -0.0001

Data Index: 153043 (C4 (Web Text))

Max Activating Token Index: 72

Text #13

Max Range: 5.7686. Min Range: -5.7686

Max Act: 3.5356. Min Act: -0.0001

Data Index: 1053528 (C4 (Web Text))

Max Activating Token Index: 349

Text #14

Max Range: 5.7686. Min Range: -5.7686

Max Act: 3.4532. Min Act: -0.0001

Data Index: 235802 (C4 (Web Text))

Max Activating Token Index: 218

Text #15

Max Range: 5.7686. Min Range: -5.7686

Max Act: 3.7145. Min Act: -0.0001

Data Index: 964176 (C4 (Web Text))

Max Activating Token Index: 647

Text #16

Max Range: 5.7686. Min Range: -5.7686

Max Act: 3.5225. Min Act: -0.0001

Data Index: 1291347 (C4 (Web Text))

Max Activating Token Index: 543

Text #17

Max Range: 5.7686. Min Range: -5.7686

Max Act: 3.3173. Min Act: -0.0001

Data Index: 51565 (C4 (Web Text))

Max Activating Token Index: 568

Text #18

Max Range: 5.7686. Min Range: -5.7686

Max Act: 3.5939. Min Act: -0.0001

Data Index: 1035309 (C4 (Web Text))

Max Activating Token Index: 928

Text #19

Max Range: 5.7686. Min Range: -5.7686

Max Act: 3.4208. Min Act: -0.0001

Data Index: 485525 (C4 (Web Text))

Max Activating Token Index: 341