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

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

Neuron 2257 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.1935. Min Range: -3.1935

Max Act: 3.1935. Min Act: -0.0001

Data Index: 347660 (C4 (Web Text))

Max Activating Token Index: 920

Text #1

Max Range: 3.1935. Min Range: -3.1935

Max Act: 3.1042. Min Act: -0.0001

Data Index: 1302892 (C4 (Web Text))

Max Activating Token Index: 72

Text #2

Max Range: 3.1935. Min Range: -3.1935

Max Act: 2.9232. Min Act: -0.0001

Data Index: 588241 (C4 (Web Text))

Max Activating Token Index: 512

Text #3

Max Range: 3.1935. Min Range: -3.1935

Max Act: 2.9896. Min Act: -0.0001

Data Index: 769846 (C4 (Web Text))

Max Activating Token Index: 460

Text #4

Max Range: 3.1935. Min Range: -3.1935

Max Act: 2.5820. Min Act: -0.0001

Data Index: 1059540 (C4 (Web Text))

Max Activating Token Index: 836

Text #5

Max Range: 3.1935. Min Range: -3.1935

Max Act: 2.6739. Min Act: -0.0001

Data Index: 1453374 (Python Code)

Max Activating Token Index: 137

Text #6

Max Range: 3.1935. Min Range: -3.1935

Max Act: 2.6701. Min Act: -0.0001

Data Index: 616502 (C4 (Web Text))

Max Activating Token Index: 764

Text #7

Max Range: 3.1935. Min Range: -3.1935

Max Act: 2.7528. Min Act: -0.0001

Data Index: 825330 (C4 (Web Text))

Max Activating Token Index: 356

Text #8

Max Range: 3.1935. Min Range: -3.1935

Max Act: 2.5856. Min Act: -0.0001

Data Index: 979752 (C4 (Web Text))

Max Activating Token Index: 99

Text #9

Max Range: 3.1935. Min Range: -3.1935

Max Act: 2.4495. Min Act: -0.0001

Data Index: 750001 (C4 (Web Text))

Max Activating Token Index: 621

Text #10

Max Range: 3.1935. Min Range: -3.1935

Max Act: 2.5474. Min Act: -0.0001

Data Index: 1600361 (Python Code)

Max Activating Token Index: 329

Text #11

Max Range: 3.1935. Min Range: -3.1935

Max Act: 2.5311. Min Act: -0.0001

Data Index: 1136727 (C4 (Web Text))

Max Activating Token Index: 273

Text #12

Max Range: 3.1935. Min Range: -3.1935

Max Act: 2.5291. Min Act: -0.0001

Data Index: 1555056 (Python Code)

Max Activating Token Index: 85

Text #13

Max Range: 3.1935. Min Range: -3.1935

Max Act: 2.4064. Min Act: -0.0001

Data Index: 84312 (C4 (Web Text))

Max Activating Token Index: 482

Text #14

Max Range: 3.1935. Min Range: -3.1935

Max Act: 2.5199. Min Act: -0.0001

Data Index: 1521821 (Python Code)

Max Activating Token Index: 250

Text #15

Max Range: 3.1935. Min Range: -3.1935

Max Act: 2.3950. Min Act: -0.0001

Data Index: 618634 (C4 (Web Text))

Max Activating Token Index: 612

Text #16

Max Range: 3.1935. Min Range: -3.1935

Max Act: 2.4721. Min Act: -0.0001

Data Index: 1642619 (Python Code)

Max Activating Token Index: 952

Text #17

Max Range: 3.1935. Min Range: -3.1935

Max Act: 2.1084. Min Act: -0.0001

Data Index: 407600 (C4 (Web Text))

Max Activating Token Index: 952

Text #18

Max Range: 3.1935. Min Range: -3.1935

Max Act: 2.4390. Min Act: -0.0001

Data Index: 1433441 (Python Code)

Max Activating Token Index: 70

Text #19

Max Range: 3.1935. Min Range: -3.1935

Max Act: 2.4422. Min Act: -0.0001

Data Index: 285417 (C4 (Web Text))

Max Activating Token Index: 785