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

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

Neuron 4417 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.4365. Min Range: -2.4365

Max Act: 2.4365. Min Act: -0.0000

Data Index: 135380 (C4 (Web Text))

Max Activating Token Index: 133

Text #1

Max Range: 2.4365. Min Range: -2.4365

Max Act: 2.3114. Min Act: -0.0000

Data Index: 330844 (C4 (Web Text))

Max Activating Token Index: 160

Text #2

Max Range: 2.4365. Min Range: -2.4365

Max Act: 2.0465. Min Act: -0.0000

Data Index: 1652435 (Python Code)

Max Activating Token Index: 367

Text #3

Max Range: 2.4365. Min Range: -2.4365

Max Act: 2.2183. Min Act: -0.0000

Data Index: 807232 (C4 (Web Text))

Max Activating Token Index: 637

Text #4

Max Range: 2.4365. Min Range: -2.4365

Max Act: 1.8964. Min Act: -0.0000

Data Index: 82388 (C4 (Web Text))

Max Activating Token Index: 136

Text #5

Max Range: 2.4365. Min Range: -2.4365

Max Act: 1.8565. Min Act: -0.0000

Data Index: 893358 (C4 (Web Text))

Max Activating Token Index: 966

Text #6

Max Range: 2.4365. Min Range: -2.4365

Max Act: 1.8186. Min Act: -0.0000

Data Index: 119322 (C4 (Web Text))

Max Activating Token Index: 975

Text #7

Max Range: 2.4365. Min Range: -2.4365

Max Act: 1.6077. Min Act: -0.0000

Data Index: 12829 (C4 (Web Text))

Max Activating Token Index: 679

Text #8

Max Range: 2.4365. Min Range: -2.4365

Max Act: 1.8500. Min Act: -0.0000

Data Index: 180878 (C4 (Web Text))

Max Activating Token Index: 732

Text #9

Max Range: 2.4365. Min Range: -2.4365

Max Act: 1.6654. Min Act: -0.0000

Data Index: 1628130 (Python Code)

Max Activating Token Index: 829

Text #10

Max Range: 2.4365. Min Range: -2.4365

Max Act: 1.5309. Min Act: -0.0000

Data Index: 6100 (C4 (Web Text))

Max Activating Token Index: 712

Text #11

Max Range: 2.4365. Min Range: -2.4365

Max Act: 1.6000. Min Act: -0.0000

Data Index: 1124904 (C4 (Web Text))

Max Activating Token Index: 638

Text #12

Max Range: 2.4365. Min Range: -2.4365

Max Act: 1.5397. Min Act: -0.0000

Data Index: 1553988 (Python Code)

Max Activating Token Index: 1003

Text #13

Max Range: 2.4365. Min Range: -2.4365

Max Act: 1.5259. Min Act: -0.0000

Data Index: 905350 (C4 (Web Text))

Max Activating Token Index: 188

Text #14

Max Range: 2.4365. Min Range: -2.4365

Max Act: 1.4782. Min Act: -0.0000

Data Index: 826734 (C4 (Web Text))

Max Activating Token Index: 403

Text #15

Max Range: 2.4365. Min Range: -2.4365

Max Act: 1.4761. Min Act: -0.0000

Data Index: 1520970 (Python Code)

Max Activating Token Index: 492

Text #16

Max Range: 2.4365. Min Range: -2.4365

Max Act: 1.3643. Min Act: -0.0000

Data Index: 1596821 (Python Code)

Max Activating Token Index: 603

Text #17

Max Range: 2.4365. Min Range: -2.4365

Max Act: 1.3759. Min Act: -0.0000

Data Index: 539265 (C4 (Web Text))

Max Activating Token Index: 229

Text #18

Max Range: 2.4365. Min Range: -2.4365

Max Act: 1.3337. Min Act: -0.0000

Data Index: 1224367 (C4 (Web Text))

Max Activating Token Index: 198

Text #19

Max Range: 2.4365. Min Range: -2.4365

Max Act: 1.4069. Min Act: -0.0000

Data Index: 1575548 (Python Code)

Max Activating Token Index: 170