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

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

Neuron 111 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1461. Min Range: -0.1461

Max Act: 0.1461. Min Act: -0.0001

Data Index: 328657 (C4 (Web Text))

Max Activating Token Index: 77

Text #1

Max Range: 0.1461. Min Range: -0.1461

Max Act: 0.1197. Min Act: -0.0001

Data Index: 1575083 (Python Code)

Max Activating Token Index: 1013

Text #2

Max Range: 0.1461. Min Range: -0.1461

Max Act: 0.1321. Min Act: -0.0001

Data Index: 1294441 (C4 (Web Text))

Max Activating Token Index: 668

Text #3

Max Range: 0.1461. Min Range: -0.1461

Max Act: 0.1180. Min Act: -0.0001

Data Index: 1655363 (Python Code)

Max Activating Token Index: 1014

Text #4

Max Range: 0.1461. Min Range: -0.1461

Max Act: 0.1036. Min Act: -0.0001

Data Index: 1417646 (Python Code)

Max Activating Token Index: 606

Text #5

Max Range: 0.1461. Min Range: -0.1461

Max Act: 0.1060. Min Act: -0.0001

Data Index: 713355 (C4 (Web Text))

Max Activating Token Index: 909

Text #6

Max Range: 0.1461. Min Range: -0.1461

Max Act: 0.1042. Min Act: -0.0001

Data Index: 777105 (C4 (Web Text))

Max Activating Token Index: 920

Text #7

Max Range: 0.1461. Min Range: -0.1461

Max Act: 0.1148. Min Act: -0.0001

Data Index: 162534 (C4 (Web Text))

Max Activating Token Index: 614

Text #8

Max Range: 0.1461. Min Range: -0.1461

Max Act: 0.1075. Min Act: -0.0001

Data Index: 1469552 (Python Code)

Max Activating Token Index: 413

Text #9

Max Range: 0.1461. Min Range: -0.1461

Max Act: 0.1093. Min Act: -0.0001

Data Index: 362386 (C4 (Web Text))

Max Activating Token Index: 397

Text #10

Max Range: 0.1461. Min Range: -0.1461

Max Act: 0.1041. Min Act: -0.0001

Data Index: 1496857 (Python Code)

Max Activating Token Index: 480

Text #11

Max Range: 0.1461. Min Range: -0.1461

Max Act: 0.0997. Min Act: -0.0001

Data Index: 1033788 (C4 (Web Text))

Max Activating Token Index: 1003

Text #12

Max Range: 0.1461. Min Range: -0.1461

Max Act: 0.1037. Min Act: -0.0001

Data Index: 1596750 (Python Code)

Max Activating Token Index: 433

Text #13

Max Range: 0.1461. Min Range: -0.1461

Max Act: 0.1023. Min Act: -0.0001

Data Index: 171580 (C4 (Web Text))

Max Activating Token Index: 814

Text #14

Max Range: 0.1461. Min Range: -0.1461

Max Act: 0.0939. Min Act: -0.0001

Data Index: 1463260 (Python Code)

Max Activating Token Index: 780

Text #15

Max Range: 0.1461. Min Range: -0.1461

Max Act: 0.1021. Min Act: -0.0001

Data Index: 373586 (C4 (Web Text))

Max Activating Token Index: 485

Text #16

Max Range: 0.1461. Min Range: -0.1461

Max Act: 0.0964. Min Act: -0.0001

Data Index: 914174 (C4 (Web Text))

Max Activating Token Index: 133

Text #17

Max Range: 0.1461. Min Range: -0.1461

Max Act: 0.0953. Min Act: -0.0001

Data Index: 1588816 (Python Code)

Max Activating Token Index: 436

Text #18

Max Range: 0.1461. Min Range: -0.1461

Max Act: 0.1052. Min Act: -0.0001

Data Index: 1288623 (C4 (Web Text))

Max Activating Token Index: 897

Text #19

Max Range: 0.1461. Min Range: -0.1461

Max Act: 0.0949. Min Act: -0.0001

Data Index: 1627947 (Python Code)

Max Activating Token Index: 326