Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 4473 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 4.6432. Min Range: -4.6432

Max Act: 4.1537. Min Act: 0.0000

Data Index: 1370708 (Python Code)

Max Activating Token Index: 387

Text #1

Max Range: 4.6432. Min Range: -4.6432

Max Act: 4.6432. Min Act: -0.0000

Data Index: 290188 (C4 (Web Text))

Max Activating Token Index: 283

Text #2

Max Range: 4.6432. Min Range: -4.6432

Max Act: 4.4189. Min Act: -0.0000

Data Index: 1214704 (C4 (Web Text))

Max Activating Token Index: 225

Text #3

Max Range: 4.6432. Min Range: -4.6432

Max Act: 3.7853. Min Act: -0.0000

Data Index: 235144 (C4 (Web Text))

Max Activating Token Index: 216

Text #4

Max Range: 4.6432. Min Range: -4.6432

Max Act: 4.2149. Min Act: -0.0000

Data Index: 1655948 (Python Code)

Max Activating Token Index: 538

Text #5

Max Range: 4.6432. Min Range: -4.6432

Max Act: 4.3819. Min Act: -0.0000

Data Index: 907703 (C4 (Web Text))

Max Activating Token Index: 200

Text #6

Max Range: 4.6432. Min Range: -4.6432

Max Act: 4.1918. Min Act: -0.0000

Data Index: 315114 (C4 (Web Text))

Max Activating Token Index: 856

Text #7

Max Range: 4.6432. Min Range: -4.6432

Max Act: 3.9553. Min Act: -0.0000

Data Index: 1561531 (Python Code)

Max Activating Token Index: 630

Text #8

Max Range: 4.6432. Min Range: -4.6432

Max Act: 3.9449. Min Act: -0.0000

Data Index: 356394 (C4 (Web Text))

Max Activating Token Index: 343

Text #9

Max Range: 4.6432. Min Range: -4.6432

Max Act: 4.1003. Min Act: 0.0000

Data Index: 1053100 (C4 (Web Text))

Max Activating Token Index: 296

Text #10

Max Range: 4.6432. Min Range: -4.6432

Max Act: 3.8854. Min Act: -0.0000

Data Index: 425762 (C4 (Web Text))

Max Activating Token Index: 300

Text #11

Max Range: 4.6432. Min Range: -4.6432

Max Act: 4.0191. Min Act: -0.0000

Data Index: 113411 (C4 (Web Text))

Max Activating Token Index: 215

Text #12

Max Range: 4.6432. Min Range: -4.6432

Max Act: 3.9903. Min Act: -0.0000

Data Index: 858813 (C4 (Web Text))

Max Activating Token Index: 213

Text #13

Max Range: 4.6432. Min Range: -4.6432

Max Act: 4.2751. Min Act: -0.0000

Data Index: 3385 (C4 (Web Text))

Max Activating Token Index: 116

Text #14

Max Range: 4.6432. Min Range: -4.6432

Max Act: 3.9022. Min Act: -0.0000

Data Index: 958823 (C4 (Web Text))

Max Activating Token Index: 165

Text #15

Max Range: 4.6432. Min Range: -4.6432

Max Act: 3.8969. Min Act: 0.0000

Data Index: 1139092 (C4 (Web Text))

Max Activating Token Index: 327

Text #16

Max Range: 4.6432. Min Range: -4.6432

Max Act: 3.8900. Min Act: 0.0000

Data Index: 342678 (C4 (Web Text))

Max Activating Token Index: 177

Text #17

Max Range: 4.6432. Min Range: -4.6432

Max Act: 3.3589. Min Act: -0.0000

Data Index: 1047004 (C4 (Web Text))

Max Activating Token Index: 694

Text #18

Max Range: 4.6432. Min Range: -4.6432

Max Act: 3.7970. Min Act: -0.0000

Data Index: 466550 (C4 (Web Text))

Max Activating Token Index: 610

Text #19

Max Range: 4.6432. Min Range: -4.6432

Max Act: 3.9439. Min Act: -0.0000

Data Index: 1484070 (Python Code)

Max Activating Token Index: 491