Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 2235 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.2897. Min Range: -0.2897

Max Act: 0.2897. Min Act: -0.0001

Data Index: 1609915 (Python Code)

Max Activating Token Index: 13

Text #1

Max Range: 0.2897. Min Range: -0.2897

Max Act: 0.2492. Min Act: -0.0001

Data Index: 1455678 (Python Code)

Max Activating Token Index: 14

Text #2

Max Range: 0.2897. Min Range: -0.2897

Max Act: 0.2458. Min Act: -0.0001

Data Index: 952226 (C4 (Web Text))

Max Activating Token Index: 852

Text #3

Max Range: 0.2897. Min Range: -0.2897

Max Act: 0.2526. Min Act: -0.0001

Data Index: 931302 (C4 (Web Text))

Max Activating Token Index: 699

Text #4

Max Range: 0.2897. Min Range: -0.2897

Max Act: 0.2143. Min Act: -0.0001

Data Index: 107188 (C4 (Web Text))

Max Activating Token Index: 901

Text #5

Max Range: 0.2897. Min Range: -0.2897

Max Act: 0.2047. Min Act: -0.0001

Data Index: 161707 (C4 (Web Text))

Max Activating Token Index: 590

Text #6

Max Range: 0.2897. Min Range: -0.2897

Max Act: 0.2471. Min Act: -0.0001

Data Index: 945864 (C4 (Web Text))

Max Activating Token Index: 901

Text #7

Max Range: 0.2897. Min Range: -0.2897

Max Act: 0.2048. Min Act: -0.0001

Data Index: 538519 (C4 (Web Text))

Max Activating Token Index: 318

Text #8

Max Range: 0.2897. Min Range: -0.2897

Max Act: 0.2188. Min Act: -0.0001

Data Index: 525618 (C4 (Web Text))

Max Activating Token Index: 910

Text #9

Max Range: 0.2897. Min Range: -0.2897

Max Act: 0.2512. Min Act: -0.0001

Data Index: 783662 (C4 (Web Text))

Max Activating Token Index: 822

Text #10

Max Range: 0.2897. Min Range: -0.2897

Max Act: 0.2146. Min Act: -0.0001

Data Index: 858858 (C4 (Web Text))

Max Activating Token Index: 498

Text #11

Max Range: 0.2897. Min Range: -0.2897

Max Act: 0.2048. Min Act: -0.0001

Data Index: 969536 (C4 (Web Text))

Max Activating Token Index: 269

Text #12

Max Range: 0.2897. Min Range: -0.2897

Max Act: 0.2389. Min Act: -0.0001

Data Index: 619068 (C4 (Web Text))

Max Activating Token Index: 621

Text #13

Max Range: 0.2897. Min Range: -0.2897

Max Act: 0.2403. Min Act: -0.0001

Data Index: 793722 (C4 (Web Text))

Max Activating Token Index: 527

Text #14

Max Range: 0.2897. Min Range: -0.2897

Max Act: 0.2323. Min Act: -0.0001

Data Index: 400913 (C4 (Web Text))

Max Activating Token Index: 695

Text #15

Max Range: 0.2897. Min Range: -0.2897

Max Act: 0.2193. Min Act: -0.0001

Data Index: 1550178 (Python Code)

Max Activating Token Index: 118

Text #16

Max Range: 0.2897. Min Range: -0.2897

Max Act: 0.2586. Min Act: -0.0001

Data Index: 1257989 (C4 (Web Text))

Max Activating Token Index: 385

Text #17

Max Range: 0.2897. Min Range: -0.2897

Max Act: 0.2273. Min Act: -0.0001

Data Index: 94558 (C4 (Web Text))

Max Activating Token Index: 931

Text #18

Max Range: 0.2897. Min Range: -0.2897

Max Act: 0.2305. Min Act: -0.0001

Data Index: 708234 (C4 (Web Text))

Max Activating Token Index: 442

Text #19

Max Range: 0.2897. Min Range: -0.2897

Max Act: 0.2435. Min Act: -0.0001

Data Index: 671133 (C4 (Web Text))

Max Activating Token Index: 862