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

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

Neuron 2785 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.1085. Min Range: -1.1085

Max Act: 0.9050. Min Act: -0.0001

Data Index: 1296937 (C4 (Web Text))

Max Activating Token Index: 573

Text #1

Max Range: 1.1085. Min Range: -1.1085

Max Act: 0.8920. Min Act: -0.0001

Data Index: 1100248 (C4 (Web Text))

Max Activating Token Index: 979

Text #2

Max Range: 1.1085. Min Range: -1.1085

Max Act: 0.9678. Min Act: -0.0001

Data Index: 1651383 (Python Code)

Max Activating Token Index: 239

Text #3

Max Range: 1.1085. Min Range: -1.1085

Max Act: 0.9350. Min Act: -0.0001

Data Index: 1413109 (Python Code)

Max Activating Token Index: 1008

Text #4

Max Range: 1.1085. Min Range: -1.1085

Max Act: 0.9493. Min Act: -0.0001

Data Index: 48872 (C4 (Web Text))

Max Activating Token Index: 583

Text #5

Max Range: 1.1085. Min Range: -1.1085

Max Act: 0.8588. Min Act: -0.0001

Data Index: 157482 (C4 (Web Text))

Max Activating Token Index: 820

Text #6

Max Range: 1.1085. Min Range: -1.1085

Max Act: 0.8599. Min Act: -0.0001

Data Index: 175746 (C4 (Web Text))

Max Activating Token Index: 319

Text #7

Max Range: 1.1085. Min Range: -1.1085

Max Act: 0.9098. Min Act: -0.0001

Data Index: 1409043 (Python Code)

Max Activating Token Index: 337

Text #8

Max Range: 1.1085. Min Range: -1.1085

Max Act: 0.9071. Min Act: -0.0001

Data Index: 1086358 (C4 (Web Text))

Max Activating Token Index: 449

Text #9

Max Range: 1.1085. Min Range: -1.1085

Max Act: 0.8978. Min Act: -0.0001

Data Index: 1594116 (Python Code)

Max Activating Token Index: 913

Text #10

Max Range: 1.1085. Min Range: -1.1085

Max Act: 1.1085. Min Act: -0.0001

Data Index: 857988 (C4 (Web Text))

Max Activating Token Index: 788

Text #11

Max Range: 1.1085. Min Range: -1.1085

Max Act: 0.8307. Min Act: -0.0001

Data Index: 1462940 (Python Code)

Max Activating Token Index: 924

Text #12

Max Range: 1.1085. Min Range: -1.1085

Max Act: 0.8315. Min Act: -0.0001

Data Index: 555373 (C4 (Web Text))

Max Activating Token Index: 543

Text #13

Max Range: 1.1085. Min Range: -1.1085

Max Act: 0.8787. Min Act: -0.0001

Data Index: 505123 (C4 (Web Text))

Max Activating Token Index: 87

Text #14

Max Range: 1.1085. Min Range: -1.1085

Max Act: 0.7956. Min Act: -0.0001

Data Index: 1268888 (C4 (Web Text))

Max Activating Token Index: 285

Text #15

Max Range: 1.1085. Min Range: -1.1085

Max Act: 0.8455. Min Act: -0.0001

Data Index: 2928 (C4 (Web Text))

Max Activating Token Index: 363

Text #16

Max Range: 1.1085. Min Range: -1.1085

Max Act: 0.7017. Min Act: -0.0001

Data Index: 307207 (C4 (Web Text))

Max Activating Token Index: 179

Text #17

Max Range: 1.1085. Min Range: -1.1085

Max Act: 0.7990. Min Act: -0.0001

Data Index: 723949 (C4 (Web Text))

Max Activating Token Index: 250

Text #18

Max Range: 1.1085. Min Range: -1.1085

Max Act: 0.8408. Min Act: -0.0001

Data Index: 418912 (C4 (Web Text))

Max Activating Token Index: 789

Text #19

Max Range: 1.1085. Min Range: -1.1085

Max Act: 0.7135. Min Act: -0.0001

Data Index: 423711 (C4 (Web Text))

Max Activating Token Index: 647