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

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

Neuron 2112 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6733. Min Range: -0.6733

Max Act: 0.6733. Min Act: -0.0000

Data Index: 845590 (C4 (Web Text))

Max Activating Token Index: 485

Text #1

Max Range: 0.6733. Min Range: -0.6733

Max Act: 0.5933. Min Act: -0.0000

Data Index: 1043115 (C4 (Web Text))

Max Activating Token Index: 237

Text #2

Max Range: 0.6733. Min Range: -0.6733

Max Act: 0.5410. Min Act: -0.0000

Data Index: 920533 (C4 (Web Text))

Max Activating Token Index: 138

Text #3

Max Range: 0.6733. Min Range: -0.6733

Max Act: 0.5819. Min Act: -0.0000

Data Index: 295402 (C4 (Web Text))

Max Activating Token Index: 963

Text #4

Max Range: 0.6733. Min Range: -0.6733

Max Act: 0.5538. Min Act: -0.0000

Data Index: 1633821 (Python Code)

Max Activating Token Index: 822

Text #5

Max Range: 0.6733. Min Range: -0.6733

Max Act: 0.4851. Min Act: -0.0000

Data Index: 1304211 (C4 (Web Text))

Max Activating Token Index: 569

Text #6

Max Range: 0.6733. Min Range: -0.6733

Max Act: 0.5356. Min Act: -0.0000

Data Index: 762414 (C4 (Web Text))

Max Activating Token Index: 550

Text #7

Max Range: 0.6733. Min Range: -0.6733

Max Act: 0.5273. Min Act: -0.0000

Data Index: 1253563 (C4 (Web Text))

Max Activating Token Index: 882

Text #8

Max Range: 0.6733. Min Range: -0.6733

Max Act: 0.5028. Min Act: -0.0001

Data Index: 1516289 (Python Code)

Max Activating Token Index: 852

Text #9

Max Range: 0.6733. Min Range: -0.6733

Max Act: 0.5161. Min Act: -0.0000

Data Index: 1159139 (C4 (Web Text))

Max Activating Token Index: 161

Text #10

Max Range: 0.6733. Min Range: -0.6733

Max Act: 0.4478. Min Act: -0.0000

Data Index: 641949 (C4 (Web Text))

Max Activating Token Index: 982

Text #11

Max Range: 0.6733. Min Range: -0.6733

Max Act: 0.5467. Min Act: -0.0000

Data Index: 890200 (C4 (Web Text))

Max Activating Token Index: 550

Text #12

Max Range: 0.6733. Min Range: -0.6733

Max Act: 0.4890. Min Act: -0.0000

Data Index: 1085160 (C4 (Web Text))

Max Activating Token Index: 887

Text #13

Max Range: 0.6733. Min Range: -0.6733

Max Act: 0.4533. Min Act: -0.0000

Data Index: 721485 (C4 (Web Text))

Max Activating Token Index: 949

Text #14

Max Range: 0.6733. Min Range: -0.6733

Max Act: 0.4341. Min Act: -0.0000

Data Index: 1198886 (C4 (Web Text))

Max Activating Token Index: 323

Text #15

Max Range: 0.6733. Min Range: -0.6733

Max Act: 0.4283. Min Act: -0.0000

Data Index: 1025171 (C4 (Web Text))

Max Activating Token Index: 415

Text #16

Max Range: 0.6733. Min Range: -0.6733

Max Act: 0.4475. Min Act: -0.0000

Data Index: 338312 (C4 (Web Text))

Max Activating Token Index: 661

Text #17

Max Range: 0.6733. Min Range: -0.6733

Max Act: 0.4154. Min Act: -0.0000

Data Index: 665348 (C4 (Web Text))

Max Activating Token Index: 364

Text #18

Max Range: 0.6733. Min Range: -0.6733

Max Act: 0.4154. Min Act: -0.0000

Data Index: 42927 (C4 (Web Text))

Max Activating Token Index: 601

Text #19

Max Range: 0.6733. Min Range: -0.6733

Max Act: 0.4242. Min Act: -0.0001

Data Index: 1485494 (Python Code)

Max Activating Token Index: 711