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

Dataset: The Pile

Neuron 4320 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 10.6280. Min Range: -10.6280

Max Act: 10.3892. Min Act: -0.0000

Data Index: 1478269 (The Pile)

Max Activating Token Index: 46

Text #1

Max Range: 10.6280. Min Range: -10.6280

Max Act: 10.6280. Min Act: -0.0000

Data Index: 1771788 (The Pile)

Max Activating Token Index: 22

Text #2

Max Range: 10.6280. Min Range: -10.6280

Max Act: 10.4569. Min Act: -0.0000

Data Index: 1669828 (The Pile)

Max Activating Token Index: 42

Text #3

Max Range: 10.6280. Min Range: -10.6280

Max Act: 10.2173. Min Act: -0.0000

Data Index: 10688 (The Pile)

Max Activating Token Index: 69

Text #4

Max Range: 10.6280. Min Range: -10.6280

Max Act: 9.8290. Min Act: -0.0000

Data Index: 987186 (The Pile)

Max Activating Token Index: 640

Text #5

Max Range: 10.6280. Min Range: -10.6280

Max Act: 9.6513. Min Act: -0.0000

Data Index: 1455042 (The Pile)

Max Activating Token Index: 534

Text #6

Max Range: 10.6280. Min Range: -10.6280

Max Act: 10.0245. Min Act: -0.0000

Data Index: 675844 (The Pile)

Max Activating Token Index: 49

Text #7

Max Range: 10.6280. Min Range: -10.6280

Max Act: 9.7312. Min Act: -0.0000

Data Index: 359399 (The Pile)

Max Activating Token Index: 40

Text #8

Max Range: 10.6280. Min Range: -10.6280

Max Act: 9.7153. Min Act: -0.0000

Data Index: 1655171 (The Pile)

Max Activating Token Index: 48

Text #9

Max Range: 10.6280. Min Range: -10.6280

Max Act: 9.8399. Min Act: -0.0000

Data Index: 234954 (The Pile)

Max Activating Token Index: 600

Text #10

Max Range: 10.6280. Min Range: -10.6280

Max Act: 9.7320. Min Act: -0.0000

Data Index: 1253357 (The Pile)

Max Activating Token Index: 814

Text #11

Max Range: 10.6280. Min Range: -10.6280

Max Act: 9.9778. Min Act: -0.0000

Data Index: 1623207 (The Pile)

Max Activating Token Index: 243

Text #12

Max Range: 10.6280. Min Range: -10.6280

Max Act: 9.8027. Min Act: -0.0000

Data Index: 1155623 (The Pile)

Max Activating Token Index: 74

Text #13

Max Range: 10.6280. Min Range: -10.6280

Max Act: 9.5436. Min Act: -0.0000

Data Index: 490596 (The Pile)

Max Activating Token Index: 276

Text #14

Max Range: 10.6280. Min Range: -10.6280

Max Act: 9.6653. Min Act: -0.0000

Data Index: 1633642 (The Pile)

Max Activating Token Index: 239

Text #15

Max Range: 10.6280. Min Range: -10.6280

Max Act: 9.8952. Min Act: -0.0000

Data Index: 1648554 (The Pile)

Max Activating Token Index: 79

Text #16

Max Range: 10.6280. Min Range: -10.6280

Max Act: 9.7497. Min Act: -0.0000

Data Index: 1610893 (The Pile)

Max Activating Token Index: 23

Text #17

Max Range: 10.6280. Min Range: -10.6280

Max Act: 9.7388. Min Act: -0.0000

Data Index: 1362931 (The Pile)

Max Activating Token Index: 20

Text #18

Max Range: 10.6280. Min Range: -10.6280

Max Act: 9.7253. Min Act: -0.0000

Data Index: 1556139 (The Pile)

Max Activating Token Index: 79

Text #19

Max Range: 10.6280. Min Range: -10.6280

Max Act: 9.6725. Min Act: -0.0000

Data Index: 126178 (The Pile)

Max Activating Token Index: 74