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

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 3206 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.0161. Min Range: -1.0161

Max Act: 1.0161. Min Act: -0.0000

Data Index: 1458999 (Python Code)

Max Activating Token Index: 673

Text #1

Max Range: 1.0161. Min Range: -1.0161

Max Act: 1.0085. Min Act: -0.0000

Data Index: 1159655 (C4 (Web Text))

Max Activating Token Index: 786

Text #2

Max Range: 1.0161. Min Range: -1.0161

Max Act: 0.8791. Min Act: -0.0000

Data Index: 270321 (C4 (Web Text))

Max Activating Token Index: 550

Text #3

Max Range: 1.0161. Min Range: -1.0161

Max Act: 0.9037. Min Act: -0.0000

Data Index: 897362 (C4 (Web Text))

Max Activating Token Index: 892

Text #4

Max Range: 1.0161. Min Range: -1.0161

Max Act: 0.8457. Min Act: -0.0000

Data Index: 84815 (C4 (Web Text))

Max Activating Token Index: 607

Text #5

Max Range: 1.0161. Min Range: -1.0161

Max Act: 0.8474. Min Act: -0.0000

Data Index: 1013857 (C4 (Web Text))

Max Activating Token Index: 535

Text #6

Max Range: 1.0161. Min Range: -1.0161

Max Act: 0.7661. Min Act: -0.0000

Data Index: 367096 (C4 (Web Text))

Max Activating Token Index: 161

Text #7

Max Range: 1.0161. Min Range: -1.0161

Max Act: 0.8175. Min Act: -0.0000

Data Index: 759742 (C4 (Web Text))

Max Activating Token Index: 855

Text #8

Max Range: 1.0161. Min Range: -1.0161

Max Act: 0.8408. Min Act: -0.0000

Data Index: 549059 (C4 (Web Text))

Max Activating Token Index: 175

Text #9

Max Range: 1.0161. Min Range: -1.0161

Max Act: 0.7997. Min Act: -0.0000

Data Index: 1342691 (C4 (Web Text))

Max Activating Token Index: 322

Text #10

Max Range: 1.0161. Min Range: -1.0161

Max Act: 0.8054. Min Act: -0.0000

Data Index: 671051 (C4 (Web Text))

Max Activating Token Index: 845

Text #11

Max Range: 1.0161. Min Range: -1.0161

Max Act: 0.8103. Min Act: -0.0001

Data Index: 910888 (C4 (Web Text))

Max Activating Token Index: 412

Text #12

Max Range: 1.0161. Min Range: -1.0161

Max Act: 0.8050. Min Act: -0.0000

Data Index: 906003 (C4 (Web Text))

Max Activating Token Index: 194

Text #13

Max Range: 1.0161. Min Range: -1.0161

Max Act: 0.8506. Min Act: -0.0000

Data Index: 403731 (C4 (Web Text))

Max Activating Token Index: 719

Text #14

Max Range: 1.0161. Min Range: -1.0161

Max Act: 0.7920. Min Act: -0.0000

Data Index: 288761 (C4 (Web Text))

Max Activating Token Index: 636

Text #15

Max Range: 1.0161. Min Range: -1.0161

Max Act: 0.7880. Min Act: -0.0000

Data Index: 95596 (C4 (Web Text))

Max Activating Token Index: 402

Text #16

Max Range: 1.0161. Min Range: -1.0161

Max Act: 0.7846. Min Act: -0.0000

Data Index: 182129 (C4 (Web Text))

Max Activating Token Index: 240

Text #17

Max Range: 1.0161. Min Range: -1.0161

Max Act: 0.7246. Min Act: -0.0000

Data Index: 26756 (C4 (Web Text))

Max Activating Token Index: 682

Text #18

Max Range: 1.0161. Min Range: -1.0161

Max Act: 0.7961. Min Act: -0.0000

Data Index: 644518 (C4 (Web Text))

Max Activating Token Index: 715

Text #19

Max Range: 1.0161. Min Range: -1.0161

Max Act: 0.7424. Min Act: -0.0000

Data Index: 624563 (C4 (Web Text))

Max Activating Token Index: 32