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

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

Neuron 2066 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5971. Min Range: -0.5971

Max Act: 0.5971. Min Act: -0.0000

Data Index: 736348 (C4 (Web Text))

Max Activating Token Index: 372

Text #1

Max Range: 0.5971. Min Range: -0.5971

Max Act: 0.5927. Min Act: -0.0000

Data Index: 1374808 (Python Code)

Max Activating Token Index: 282

Text #2

Max Range: 0.5971. Min Range: -0.5971

Max Act: 0.5968. Min Act: -0.0000

Data Index: 1414082 (Python Code)

Max Activating Token Index: 653

Text #3

Max Range: 0.5971. Min Range: -0.5971

Max Act: 0.5775. Min Act: -0.0000

Data Index: 1581170 (Python Code)

Max Activating Token Index: 698

Text #4

Max Range: 0.5971. Min Range: -0.5971

Max Act: 0.5177. Min Act: -0.0000

Data Index: 1381148 (Python Code)

Max Activating Token Index: 589

Text #5

Max Range: 0.5971. Min Range: -0.5971

Max Act: 0.5371. Min Act: -0.0000

Data Index: 1512923 (Python Code)

Max Activating Token Index: 333

Text #6

Max Range: 0.5971. Min Range: -0.5971

Max Act: 0.5005. Min Act: -0.0000

Data Index: 620077 (C4 (Web Text))

Max Activating Token Index: 320

Text #7

Max Range: 0.5971. Min Range: -0.5971

Max Act: 0.5171. Min Act: -0.0000

Data Index: 128256 (C4 (Web Text))

Max Activating Token Index: 143

Text #8

Max Range: 0.5971. Min Range: -0.5971

Max Act: 0.5172. Min Act: -0.0000

Data Index: 937353 (C4 (Web Text))

Max Activating Token Index: 870

Text #9

Max Range: 0.5971. Min Range: -0.5971

Max Act: 0.5102. Min Act: -0.0000

Data Index: 706228 (C4 (Web Text))

Max Activating Token Index: 260

Text #10

Max Range: 0.5971. Min Range: -0.5971

Max Act: 0.4855. Min Act: -0.0000

Data Index: 1587683 (Python Code)

Max Activating Token Index: 840

Text #11

Max Range: 0.5971. Min Range: -0.5971

Max Act: 0.4920. Min Act: -0.0000

Data Index: 1511596 (Python Code)

Max Activating Token Index: 249

Text #12

Max Range: 0.5971. Min Range: -0.5971

Max Act: 0.4908. Min Act: -0.0000

Data Index: 1460380 (Python Code)

Max Activating Token Index: 947

Text #13

Max Range: 0.5971. Min Range: -0.5971

Max Act: 0.4907. Min Act: -0.0000

Data Index: 850471 (C4 (Web Text))

Max Activating Token Index: 592

Text #14

Max Range: 0.5971. Min Range: -0.5971

Max Act: 0.4867. Min Act: -0.0000

Data Index: 1276487 (C4 (Web Text))

Max Activating Token Index: 438

Text #15

Max Range: 0.5971. Min Range: -0.5971

Max Act: 0.5390. Min Act: -0.0000

Data Index: 36384 (C4 (Web Text))

Max Activating Token Index: 855

Text #16

Max Range: 0.5971. Min Range: -0.5971

Max Act: 0.5059. Min Act: -0.0000

Data Index: 207595 (C4 (Web Text))

Max Activating Token Index: 902

Text #17

Max Range: 0.5971. Min Range: -0.5971

Max Act: 0.5035. Min Act: -0.0000

Data Index: 458192 (C4 (Web Text))

Max Activating Token Index: 165

Text #18

Max Range: 0.5971. Min Range: -0.5971

Max Act: 0.4803. Min Act: -0.0000

Data Index: 1366643 (Python Code)

Max Activating Token Index: 772

Text #19

Max Range: 0.5971. Min Range: -0.5971

Max Act: 0.4598. Min Act: -0.0000

Data Index: 367872 (C4 (Web Text))

Max Activating Token Index: 572