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

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

Neuron 2654 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2896. Min Range: -0.2896

Max Act: 0.2818. Min Act: -0.0001

Data Index: 440017 (C4 (Web Text))

Max Activating Token Index: 779

Text #1

Max Range: 0.2896. Min Range: -0.2896

Max Act: 0.2663. Min Act: -0.0001

Data Index: 364599 (C4 (Web Text))

Max Activating Token Index: 242

Text #2

Max Range: 0.2896. Min Range: -0.2896

Max Act: 0.2637. Min Act: -0.0001

Data Index: 533164 (C4 (Web Text))

Max Activating Token Index: 98

Text #3

Max Range: 0.2896. Min Range: -0.2896

Max Act: 0.2896. Min Act: -0.0001

Data Index: 669064 (C4 (Web Text))

Max Activating Token Index: 896

Text #4

Max Range: 0.2896. Min Range: -0.2896

Max Act: 0.2791. Min Act: -0.0001

Data Index: 389490 (C4 (Web Text))

Max Activating Token Index: 983

Text #5

Max Range: 0.2896. Min Range: -0.2896

Max Act: 0.2701. Min Act: -0.0001

Data Index: 1048297 (C4 (Web Text))

Max Activating Token Index: 996

Text #6

Max Range: 0.2896. Min Range: -0.2896

Max Act: 0.1945. Min Act: -0.0000

Data Index: 1379001 (Python Code)

Max Activating Token Index: 325

Text #7

Max Range: 0.2896. Min Range: -0.2896

Max Act: 0.2346. Min Act: -0.0000

Data Index: 1648936 (Python Code)

Max Activating Token Index: 171

Text #8

Max Range: 0.2896. Min Range: -0.2896

Max Act: 0.1976. Min Act: -0.0000

Data Index: 1361336 (Python Code)

Max Activating Token Index: 234

Text #9

Max Range: 0.2896. Min Range: -0.2896

Max Act: 0.2148. Min Act: -0.0001

Data Index: 658152 (C4 (Web Text))

Max Activating Token Index: 555

Text #10

Max Range: 0.2896. Min Range: -0.2896

Max Act: 0.2244. Min Act: -0.0000

Data Index: 1381455 (Python Code)

Max Activating Token Index: 765

Text #11

Max Range: 0.2896. Min Range: -0.2896

Max Act: 0.2099. Min Act: -0.0000

Data Index: 1400717 (Python Code)

Max Activating Token Index: 851

Text #12

Max Range: 0.2896. Min Range: -0.2896

Max Act: 0.2048. Min Act: -0.0000

Data Index: 1534754 (Python Code)

Max Activating Token Index: 996

Text #13

Max Range: 0.2896. Min Range: -0.2896

Max Act: 0.1981. Min Act: -0.0000

Data Index: 1422218 (Python Code)

Max Activating Token Index: 482

Text #14

Max Range: 0.2896. Min Range: -0.2896

Max Act: 0.2206. Min Act: -0.0000

Data Index: 1593208 (Python Code)

Max Activating Token Index: 303

Text #15

Max Range: 0.2896. Min Range: -0.2896

Max Act: 0.1861. Min Act: -0.0001

Data Index: 28831 (C4 (Web Text))

Max Activating Token Index: 867

Text #16

Max Range: 0.2896. Min Range: -0.2896

Max Act: 0.2035. Min Act: -0.0000

Data Index: 1592170 (Python Code)

Max Activating Token Index: 445

Text #17

Max Range: 0.2896. Min Range: -0.2896

Max Act: 0.1982. Min Act: -0.0000

Data Index: 1420336 (Python Code)

Max Activating Token Index: 276

Text #18

Max Range: 0.2896. Min Range: -0.2896

Max Act: 0.2044. Min Act: -0.0000

Data Index: 1431643 (Python Code)

Max Activating Token Index: 1014

Text #19

Max Range: 0.2896. Min Range: -0.2896

Max Act: 0.1934. Min Act: -0.0001

Data Index: 546240 (C4 (Web Text))

Max Activating Token Index: 17