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

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

Neuron 875 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: 1.8987. Min Range: -1.8987

Max Act: 1.8987. Min Act: -0.0001

Data Index: 1369457 (Python Code)

Max Activating Token Index: 101

Text #1

Max Range: 1.8987. Min Range: -1.8987

Max Act: 1.7644. Min Act: -0.0001

Data Index: 1466854 (Python Code)

Max Activating Token Index: 879

Text #2

Max Range: 1.8987. Min Range: -1.8987

Max Act: 1.6065. Min Act: -0.0001

Data Index: 45101 (C4 (Web Text))

Max Activating Token Index: 205

Text #3

Max Range: 1.8987. Min Range: -1.8987

Max Act: 1.5770. Min Act: -0.0001

Data Index: 742052 (C4 (Web Text))

Max Activating Token Index: 524

Text #4

Max Range: 1.8987. Min Range: -1.8987

Max Act: 1.6026. Min Act: -0.0001

Data Index: 1037010 (C4 (Web Text))

Max Activating Token Index: 953

Text #5

Max Range: 1.8987. Min Range: -1.8987

Max Act: 1.4543. Min Act: -0.0001

Data Index: 612084 (C4 (Web Text))

Max Activating Token Index: 856

Text #6

Max Range: 1.8987. Min Range: -1.8987

Max Act: 1.5818. Min Act: -0.0001

Data Index: 240204 (C4 (Web Text))

Max Activating Token Index: 616

Text #7

Max Range: 1.8987. Min Range: -1.8987

Max Act: 1.6643. Min Act: -0.0001

Data Index: 1133712 (C4 (Web Text))

Max Activating Token Index: 839

Text #8

Max Range: 1.8987. Min Range: -1.8987

Max Act: 1.4861. Min Act: -0.0001

Data Index: 451609 (C4 (Web Text))

Max Activating Token Index: 888

Text #9

Max Range: 1.8987. Min Range: -1.8987

Max Act: 1.5083. Min Act: -0.0001

Data Index: 695383 (C4 (Web Text))

Max Activating Token Index: 583

Text #10

Max Range: 1.8987. Min Range: -1.8987

Max Act: 1.6315. Min Act: -0.0001

Data Index: 16672 (C4 (Web Text))

Max Activating Token Index: 885

Text #11

Max Range: 1.8987. Min Range: -1.8987

Max Act: 1.5869. Min Act: -0.0001

Data Index: 395630 (C4 (Web Text))

Max Activating Token Index: 801

Text #12

Max Range: 1.8987. Min Range: -1.8987

Max Act: 1.4577. Min Act: -0.0001

Data Index: 1245121 (C4 (Web Text))

Max Activating Token Index: 467

Text #13

Max Range: 1.8987. Min Range: -1.8987

Max Act: 1.5327. Min Act: -0.0001

Data Index: 993513 (C4 (Web Text))

Max Activating Token Index: 268

Text #14

Max Range: 1.8987. Min Range: -1.8987

Max Act: 1.4071. Min Act: -0.0001

Data Index: 753623 (C4 (Web Text))

Max Activating Token Index: 331

Text #15

Max Range: 1.8987. Min Range: -1.8987

Max Act: 1.5198. Min Act: -0.0001

Data Index: 1273229 (C4 (Web Text))

Max Activating Token Index: 922

Text #16

Max Range: 1.8987. Min Range: -1.8987

Max Act: 1.5527. Min Act: -0.0001

Data Index: 600468 (C4 (Web Text))

Max Activating Token Index: 54

Text #17

Max Range: 1.8987. Min Range: -1.8987

Max Act: 1.5520. Min Act: -0.0001

Data Index: 1354240 (C4 (Web Text))

Max Activating Token Index: 914

Text #18

Max Range: 1.8987. Min Range: -1.8987

Max Act: 1.5128. Min Act: -0.0001

Data Index: 820010 (C4 (Web Text))

Max Activating Token Index: 865

Text #19

Max Range: 1.8987. Min Range: -1.8987

Max Act: 1.4595. Min Act: -0.0001

Data Index: 1646629 (Python Code)

Max Activating Token Index: 892