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

Dataset: The Pile

Neuron 488 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9464. Min Range: -0.9464

Max Act: 0.9464. Min Act: -0.0000

Data Index: 316979 (The Pile)

Max Activating Token Index: 948

Text #1

Max Range: 0.9464. Min Range: -0.9464

Max Act: 0.8378. Min Act: -0.0000

Data Index: 142324 (The Pile)

Max Activating Token Index: 207

Text #2

Max Range: 0.9464. Min Range: -0.9464

Max Act: 0.8049. Min Act: -0.0000

Data Index: 849172 (The Pile)

Max Activating Token Index: 846

Text #3

Max Range: 0.9464. Min Range: -0.9464

Max Act: 0.7957. Min Act: -0.0000

Data Index: 994224 (The Pile)

Max Activating Token Index: 450

Text #4

Max Range: 0.9464. Min Range: -0.9464

Max Act: 0.7970. Min Act: -0.0000

Data Index: 910169 (The Pile)

Max Activating Token Index: 263

Text #5

Max Range: 0.9464. Min Range: -0.9464

Max Act: 0.7840. Min Act: -0.0000

Data Index: 416458 (The Pile)

Max Activating Token Index: 704

Text #6

Max Range: 0.9464. Min Range: -0.9464

Max Act: 0.7603. Min Act: -0.0000

Data Index: 1948360 (The Pile)

Max Activating Token Index: 108

Text #7

Max Range: 0.9464. Min Range: -0.9464

Max Act: 0.7290. Min Act: -0.0000

Data Index: 1287017 (The Pile)

Max Activating Token Index: 498

Text #8

Max Range: 0.9464. Min Range: -0.9464

Max Act: 0.7864. Min Act: -0.0000

Data Index: 1133181 (The Pile)

Max Activating Token Index: 497

Text #9

Max Range: 0.9464. Min Range: -0.9464

Max Act: 0.7665. Min Act: -0.0000

Data Index: 1925212 (The Pile)

Max Activating Token Index: 171

Text #10

Max Range: 0.9464. Min Range: -0.9464

Max Act: 0.7559. Min Act: -0.0000

Data Index: 549368 (The Pile)

Max Activating Token Index: 876

Text #11

Max Range: 0.9464. Min Range: -0.9464

Max Act: 0.7452. Min Act: -0.0000

Data Index: 1430391 (The Pile)

Max Activating Token Index: 428

Text #12

Max Range: 0.9464. Min Range: -0.9464

Max Act: 0.7633. Min Act: -0.0000

Data Index: 1089606 (The Pile)

Max Activating Token Index: 972

Text #13

Max Range: 0.9464. Min Range: -0.9464

Max Act: 0.7122. Min Act: -0.0000

Data Index: 1491181 (The Pile)

Max Activating Token Index: 128

Text #14

Max Range: 0.9464. Min Range: -0.9464

Max Act: 0.7525. Min Act: -0.0000

Data Index: 1493070 (The Pile)

Max Activating Token Index: 469

Text #15

Max Range: 0.9464. Min Range: -0.9464

Max Act: 0.7477. Min Act: -0.0000

Data Index: 554809 (The Pile)

Max Activating Token Index: 441

Text #16

Max Range: 0.9464. Min Range: -0.9464

Max Act: 0.7237. Min Act: -0.0000

Data Index: 296992 (The Pile)

Max Activating Token Index: 73

Text #17

Max Range: 0.9464. Min Range: -0.9464

Max Act: 0.6981. Min Act: -0.0000

Data Index: 1258790 (The Pile)

Max Activating Token Index: 230

Text #18

Max Range: 0.9464. Min Range: -0.9464

Max Act: 0.7024. Min Act: -0.0000

Data Index: 136483 (The Pile)

Max Activating Token Index: 466

Text #19

Max Range: 0.9464. Min Range: -0.9464

Max Act: 0.6751. Min Act: -0.0000

Data Index: 979102 (The Pile)

Max Activating Token Index: 870