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

Dataset: The Pile

Neuron 3985 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.7741. Min Range: -1.7741

Max Act: 1.7160. Min Act: -0.0000

Data Index: 1748455 (The Pile)

Max Activating Token Index: 73

Text #1

Max Range: 1.7741. Min Range: -1.7741

Max Act: 1.7741. Min Act: -0.0000

Data Index: 1208592 (The Pile)

Max Activating Token Index: 282

Text #2

Max Range: 1.7741. Min Range: -1.7741

Max Act: 1.5603. Min Act: -0.0000

Data Index: 1228458 (The Pile)

Max Activating Token Index: 535

Text #3

Max Range: 1.7741. Min Range: -1.7741

Max Act: 1.6033. Min Act: -0.0000

Data Index: 166930 (The Pile)

Max Activating Token Index: 659

Text #4

Max Range: 1.7741. Min Range: -1.7741

Max Act: 1.6000. Min Act: -0.0000

Data Index: 14436 (The Pile)

Max Activating Token Index: 356

Text #5

Max Range: 1.7741. Min Range: -1.7741

Max Act: 1.5599. Min Act: -0.0000

Data Index: 1448554 (The Pile)

Max Activating Token Index: 642

Text #6

Max Range: 1.7741. Min Range: -1.7741

Max Act: 1.3307. Min Act: -0.0000

Data Index: 1776566 (The Pile)

Max Activating Token Index: 123

Text #7

Max Range: 1.7741. Min Range: -1.7741

Max Act: 1.4231. Min Act: -0.0000

Data Index: 384239 (The Pile)

Max Activating Token Index: 907

Text #8

Max Range: 1.7741. Min Range: -1.7741

Max Act: 1.4542. Min Act: -0.0000

Data Index: 220281 (The Pile)

Max Activating Token Index: 317

Text #9

Max Range: 1.7741. Min Range: -1.7741

Max Act: 1.4393. Min Act: -0.0000

Data Index: 1977074 (The Pile)

Max Activating Token Index: 343

Text #10

Max Range: 1.7741. Min Range: -1.7741

Max Act: 1.4015. Min Act: -0.0000

Data Index: 435231 (The Pile)

Max Activating Token Index: 25

Text #11

Max Range: 1.7741. Min Range: -1.7741

Max Act: 1.4166. Min Act: -0.0000

Data Index: 938985 (The Pile)

Max Activating Token Index: 560

Text #12

Max Range: 1.7741. Min Range: -1.7741

Max Act: 1.4933. Min Act: -0.0000

Data Index: 1901092 (The Pile)

Max Activating Token Index: 347

Text #13

Max Range: 1.7741. Min Range: -1.7741

Max Act: 1.4514. Min Act: -0.0000

Data Index: 770548 (The Pile)

Max Activating Token Index: 415

Text #14

Max Range: 1.7741. Min Range: -1.7741

Max Act: 1.4588. Min Act: -0.0000

Data Index: 1576386 (The Pile)

Max Activating Token Index: 735

Text #15

Max Range: 1.7741. Min Range: -1.7741

Max Act: 1.4463. Min Act: -0.0000

Data Index: 518817 (The Pile)

Max Activating Token Index: 299

Text #16

Max Range: 1.7741. Min Range: -1.7741

Max Act: 1.3304. Min Act: -0.0000

Data Index: 10578 (The Pile)

Max Activating Token Index: 839

Text #17

Max Range: 1.7741. Min Range: -1.7741

Max Act: 1.4785. Min Act: -0.0000

Data Index: 889042 (The Pile)

Max Activating Token Index: 296

Text #18

Max Range: 1.7741. Min Range: -1.7741

Max Act: 1.3169. Min Act: -0.0000

Data Index: 1904505 (The Pile)

Max Activating Token Index: 669

Text #19

Max Range: 1.7741. Min Range: -1.7741

Max Act: 1.4596. Min Act: -0.0000

Data Index: 440279 (The Pile)

Max Activating Token Index: 783