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

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

Neuron 960 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: 0.4641. Min Range: -0.4641

Max Act: 0.4641. Min Act: -0.0001

Data Index: 1538623 (Python Code)

Max Activating Token Index: 669

Text #1

Max Range: 0.4641. Min Range: -0.4641

Max Act: 0.4130. Min Act: 0.0000

Data Index: 393339 (C4 (Web Text))

Max Activating Token Index: 1022

Text #2

Max Range: 0.4641. Min Range: -0.4641

Max Act: 0.3992. Min Act: -0.0000

Data Index: 708034 (C4 (Web Text))

Max Activating Token Index: 69

Text #3

Max Range: 0.4641. Min Range: -0.4641

Max Act: 0.3766. Min Act: -0.0001

Data Index: 297519 (C4 (Web Text))

Max Activating Token Index: 300

Text #4

Max Range: 0.4641. Min Range: -0.4641

Max Act: 0.3608. Min Act: -0.0000

Data Index: 767128 (C4 (Web Text))

Max Activating Token Index: 963

Text #5

Max Range: 0.4641. Min Range: -0.4641

Max Act: 0.3585. Min Act: -0.0001

Data Index: 782910 (C4 (Web Text))

Max Activating Token Index: 709

Text #6

Max Range: 0.4641. Min Range: -0.4641

Max Act: 0.3622. Min Act: -0.0001

Data Index: 503244 (C4 (Web Text))

Max Activating Token Index: 169

Text #7

Max Range: 0.4641. Min Range: -0.4641

Max Act: 0.3239. Min Act: -0.0000

Data Index: 1644173 (Python Code)

Max Activating Token Index: 240

Text #8

Max Range: 0.4641. Min Range: -0.4641

Max Act: 0.3261. Min Act: -0.0001

Data Index: 209990 (C4 (Web Text))

Max Activating Token Index: 906

Text #9

Max Range: 0.4641. Min Range: -0.4641

Max Act: 0.3341. Min Act: -0.0001

Data Index: 99365 (C4 (Web Text))

Max Activating Token Index: 842

Text #10

Max Range: 0.4641. Min Range: -0.4641

Max Act: 0.3302. Min Act: -0.0000

Data Index: 425476 (C4 (Web Text))

Max Activating Token Index: 460

Text #11

Max Range: 0.4641. Min Range: -0.4641

Max Act: 0.3272. Min Act: -0.0001

Data Index: 298984 (C4 (Web Text))

Max Activating Token Index: 870

Text #12

Max Range: 0.4641. Min Range: -0.4641

Max Act: 0.3067. Min Act: -0.0000

Data Index: 370437 (C4 (Web Text))

Max Activating Token Index: 746

Text #13

Max Range: 0.4641. Min Range: -0.4641

Max Act: 0.3271. Min Act: -0.0001

Data Index: 44684 (C4 (Web Text))

Max Activating Token Index: 137

Text #14

Max Range: 0.4641. Min Range: -0.4641

Max Act: 0.3036. Min Act: -0.0001

Data Index: 1047941 (C4 (Web Text))

Max Activating Token Index: 103

Text #15

Max Range: 0.4641. Min Range: -0.4641

Max Act: 0.3120. Min Act: -0.0000

Data Index: 29985 (C4 (Web Text))

Max Activating Token Index: 1010

Text #16

Max Range: 0.4641. Min Range: -0.4641

Max Act: 0.2952. Min Act: -0.0001

Data Index: 272464 (C4 (Web Text))

Max Activating Token Index: 252

Text #17

Max Range: 0.4641. Min Range: -0.4641

Max Act: 0.3057. Min Act: -0.0001

Data Index: 166233 (C4 (Web Text))

Max Activating Token Index: 320

Text #18

Max Range: 0.4641. Min Range: -0.4641

Max Act: 0.3012. Min Act: -0.0001

Data Index: 483455 (C4 (Web Text))

Max Activating Token Index: 316

Text #19

Max Range: 0.4641. Min Range: -0.4641

Max Act: 0.3063. Min Act: -0.0000

Data Index: 1249225 (C4 (Web Text))

Max Activating Token Index: 396