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

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

Neuron 1634 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4397. Min Range: -0.4397

Max Act: 0.4397. Min Act: -0.0003

Data Index: 144449 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 0.4397. Min Range: -0.4397

Max Act: 0.4114. Min Act: -0.0003

Data Index: 1383783 (Python Code)

Max Activating Token Index: 1023

Text #2

Max Range: 0.4397. Min Range: -0.4397

Max Act: 0.3909. Min Act: -0.0003

Data Index: 477463 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 0.4397. Min Range: -0.4397

Max Act: 0.3546. Min Act: -0.0003

Data Index: 374844 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 0.4397. Min Range: -0.4397

Max Act: 0.3522. Min Act: -0.0003

Data Index: 304609 (C4 (Web Text))

Max Activating Token Index: 1023

Text #5

Max Range: 0.4397. Min Range: -0.4397

Max Act: 0.3529. Min Act: -0.0003

Data Index: 313513 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 0.4397. Min Range: -0.4397

Max Act: 0.3399. Min Act: -0.0003

Data Index: 1004685 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 0.4397. Min Range: -0.4397

Max Act: 0.3455. Min Act: -0.0003

Data Index: 1547106 (Python Code)

Max Activating Token Index: 1023

Text #8

Max Range: 0.4397. Min Range: -0.4397

Max Act: 0.3449. Min Act: -0.0003

Data Index: 1592462 (Python Code)

Max Activating Token Index: 1023

Text #9

Max Range: 0.4397. Min Range: -0.4397

Max Act: 0.3401. Min Act: -0.0003

Data Index: 251548 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 0.4397. Min Range: -0.4397

Max Act: 0.3364. Min Act: -0.0003

Data Index: 1495130 (Python Code)

Max Activating Token Index: 1023

Text #11

Max Range: 0.4397. Min Range: -0.4397

Max Act: 0.3058. Min Act: -0.0003

Data Index: 870503 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 0.4397. Min Range: -0.4397

Max Act: 0.3150. Min Act: -0.0003

Data Index: 1650535 (Python Code)

Max Activating Token Index: 1023

Text #13

Max Range: 0.4397. Min Range: -0.4397

Max Act: 0.3088. Min Act: -0.0003

Data Index: 1589378 (Python Code)

Max Activating Token Index: 1023

Text #14

Max Range: 0.4397. Min Range: -0.4397

Max Act: 0.3026. Min Act: -0.0003

Data Index: 1480109 (Python Code)

Max Activating Token Index: 1023

Text #15

Max Range: 0.4397. Min Range: -0.4397

Max Act: 0.3300. Min Act: -0.0003

Data Index: 1388579 (Python Code)

Max Activating Token Index: 1023

Text #16

Max Range: 0.4397. Min Range: -0.4397

Max Act: 0.3096. Min Act: -0.0003

Data Index: 1497640 (Python Code)

Max Activating Token Index: 1023

Text #17

Max Range: 0.4397. Min Range: -0.4397

Max Act: 0.2985. Min Act: -0.0003

Data Index: 1132729 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 0.4397. Min Range: -0.4397

Max Act: 0.2995. Min Act: -0.0003

Data Index: 1445837 (Python Code)

Max Activating Token Index: 1023

Text #19

Max Range: 0.4397. Min Range: -0.4397

Max Act: 0.3082. Min Act: -0.0003

Data Index: 1364642 (Python Code)

Max Activating Token Index: 1023