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

Dataset: The Pile

Neuron 1209 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 10.7210. Min Range: -10.7210

Max Act: 10.7210. Min Act: -0.0000

Data Index: 187821 (The Pile)

Max Activating Token Index: 920

Text #1

Max Range: 10.7210. Min Range: -10.7210

Max Act: 7.1094. Min Act: -0.0001

Data Index: 502731 (The Pile)

Max Activating Token Index: 438

Text #2

Max Range: 10.7210. Min Range: -10.7210

Max Act: 8.5018. Min Act: 0.0000

Data Index: 944803 (The Pile)

Max Activating Token Index: 186

Text #3

Max Range: 10.7210. Min Range: -10.7210

Max Act: 8.2266. Min Act: 0.0000

Data Index: 1188001 (The Pile)

Max Activating Token Index: 200

Text #4

Max Range: 10.7210. Min Range: -10.7210

Max Act: 7.3706. Min Act: -0.0000

Data Index: 1526821 (The Pile)

Max Activating Token Index: 487

Text #5

Max Range: 10.7210. Min Range: -10.7210

Max Act: 5.6128. Min Act: 0.0000

Data Index: 493308 (The Pile)

Max Activating Token Index: 70

Text #6

Max Range: 10.7210. Min Range: -10.7210

Max Act: 6.5289. Min Act: -0.0000

Data Index: 538735 (The Pile)

Max Activating Token Index: 1013

Text #7

Max Range: 10.7210. Min Range: -10.7210

Max Act: 7.1178. Min Act: 0.0000

Data Index: 94930 (The Pile)

Max Activating Token Index: 63

Text #8

Max Range: 10.7210. Min Range: -10.7210

Max Act: 5.2029. Min Act: -0.0000

Data Index: 785689 (The Pile)

Max Activating Token Index: 870

Text #9

Max Range: 10.7210. Min Range: -10.7210

Max Act: 4.8250. Min Act: -0.0001

Data Index: 1712571 (The Pile)

Max Activating Token Index: 631

Text #10

Max Range: 10.7210. Min Range: -10.7210

Max Act: 5.3988. Min Act: -0.0000

Data Index: 1011126 (The Pile)

Max Activating Token Index: 1019

Text #11

Max Range: 10.7210. Min Range: -10.7210

Max Act: 5.6535. Min Act: -0.0000

Data Index: 1925158 (The Pile)

Max Activating Token Index: 147

Text #12

Max Range: 10.7210. Min Range: -10.7210

Max Act: 5.9177. Min Act: 0.0001

Data Index: 849182 (The Pile)

Max Activating Token Index: 321

Text #13

Max Range: 10.7210. Min Range: -10.7210

Max Act: 4.7497. Min Act: -0.0001

Data Index: 282536 (The Pile)

Max Activating Token Index: 530

Text #14

Max Range: 10.7210. Min Range: -10.7210

Max Act: 6.6021. Min Act: -0.0001

Data Index: 112365 (The Pile)

Max Activating Token Index: 27

Text #15

Max Range: 10.7210. Min Range: -10.7210

Max Act: 4.8144. Min Act: -0.0000

Data Index: 153514 (The Pile)

Max Activating Token Index: 360

Text #16

Max Range: 10.7210. Min Range: -10.7210

Max Act: 4.6689. Min Act: -0.0001

Data Index: 1927301 (The Pile)

Max Activating Token Index: 747

Text #17

Max Range: 10.7210. Min Range: -10.7210

Max Act: 7.0864. Min Act: -0.0001

Data Index: 1879148 (The Pile)

Max Activating Token Index: 167

Text #18

Max Range: 10.7210. Min Range: -10.7210

Max Act: 5.9054. Min Act: -0.0000

Data Index: 167704 (The Pile)

Max Activating Token Index: 38

Text #19

Max Range: 10.7210. Min Range: -10.7210

Max Act: 7.1657. Min Act: 0.0001

Data Index: 1646339 (The Pile)

Max Activating Token Index: 211