Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

Dataset: The Pile

Neuron 2887 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l-pile')

Text #0

Max Range: 9.1890. Min Range: -9.1890

Max Act: 9.1890. Min Act: -0.0000

Data Index: 1223141 (The Pile)

Max Activating Token Index: 100

Text #1

Max Range: 9.1890. Min Range: -9.1890

Max Act: 8.7435. Min Act: -0.0000

Data Index: 485785 (The Pile)

Max Activating Token Index: 54

Text #2

Max Range: 9.1890. Min Range: -9.1890

Max Act: 8.5401. Min Act: -0.0000

Data Index: 642530 (The Pile)

Max Activating Token Index: 354

Text #3

Max Range: 9.1890. Min Range: -9.1890

Max Act: 8.1176. Min Act: -0.0000

Data Index: 580774 (The Pile)

Max Activating Token Index: 118

Text #4

Max Range: 9.1890. Min Range: -9.1890

Max Act: 8.2860. Min Act: -0.0000

Data Index: 1712413 (The Pile)

Max Activating Token Index: 141

Text #5

Max Range: 9.1890. Min Range: -9.1890

Max Act: 8.4502. Min Act: -0.0000

Data Index: 1957784 (The Pile)

Max Activating Token Index: 222

Text #6

Max Range: 9.1890. Min Range: -9.1890

Max Act: 8.8745. Min Act: -0.0000

Data Index: 805439 (The Pile)

Max Activating Token Index: 116

Text #7

Max Range: 9.1890. Min Range: -9.1890

Max Act: 8.3536. Min Act: -0.0000

Data Index: 626039 (The Pile)

Max Activating Token Index: 712

Text #8

Max Range: 9.1890. Min Range: -9.1890

Max Act: 8.1876. Min Act: -0.0000

Data Index: 1982506 (The Pile)

Max Activating Token Index: 108

Text #9

Max Range: 9.1890. Min Range: -9.1890

Max Act: 8.0709. Min Act: -0.0000

Data Index: 286560 (The Pile)

Max Activating Token Index: 182

Text #10

Max Range: 9.1890. Min Range: -9.1890

Max Act: 7.7118. Min Act: -0.0000

Data Index: 1862284 (The Pile)

Max Activating Token Index: 80

Text #11

Max Range: 9.1890. Min Range: -9.1890

Max Act: 8.1736. Min Act: -0.0000

Data Index: 632939 (The Pile)

Max Activating Token Index: 510

Text #12

Max Range: 9.1890. Min Range: -9.1890

Max Act: 7.8456. Min Act: -0.0000

Data Index: 256537 (The Pile)

Max Activating Token Index: 856

Text #13

Max Range: 9.1890. Min Range: -9.1890

Max Act: 7.9445. Min Act: -0.0000

Data Index: 571179 (The Pile)

Max Activating Token Index: 405

Text #14

Max Range: 9.1890. Min Range: -9.1890

Max Act: 5.2975. Min Act: -0.0000

Data Index: 276761 (The Pile)

Max Activating Token Index: 421

Text #15

Max Range: 9.1890. Min Range: -9.1890

Max Act: 6.6356. Min Act: -0.0000

Data Index: 841209 (The Pile)

Max Activating Token Index: 276

Text #16

Max Range: 9.1890. Min Range: -9.1890

Max Act: 7.5539. Min Act: -0.0000

Data Index: 1009647 (The Pile)

Max Activating Token Index: 478

Text #17

Max Range: 9.1890. Min Range: -9.1890

Max Act: 7.8800. Min Act: -0.0000

Data Index: 1363118 (The Pile)

Max Activating Token Index: 734

Text #18

Max Range: 9.1890. Min Range: -9.1890

Max Act: 8.2315. Min Act: -0.0000

Data Index: 1392641 (The Pile)

Max Activating Token Index: 185

Text #19

Max Range: 9.1890. Min Range: -9.1890

Max Act: 7.5217. Min Act: -0.0000

Data Index: 1538605 (The Pile)

Max Activating Token Index: 231