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

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

Neuron 581 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: 0.8104. Min Range: -0.8104

Max Act: 0.8104. Min Act: -0.0001

Data Index: 984377 (C4 (Web Text))

Max Activating Token Index: 521

Text #1

Max Range: 0.8104. Min Range: -0.8104

Max Act: 0.6327. Min Act: -0.0001

Data Index: 53548 (C4 (Web Text))

Max Activating Token Index: 924

Text #2

Max Range: 0.8104. Min Range: -0.8104

Max Act: 0.7050. Min Act: -0.0001

Data Index: 244303 (C4 (Web Text))

Max Activating Token Index: 930

Text #3

Max Range: 0.8104. Min Range: -0.8104

Max Act: 0.6096. Min Act: -0.0001

Data Index: 574539 (C4 (Web Text))

Max Activating Token Index: 981

Text #4

Max Range: 0.8104. Min Range: -0.8104

Max Act: 0.5326. Min Act: -0.0001

Data Index: 1192526 (C4 (Web Text))

Max Activating Token Index: 149

Text #5

Max Range: 0.8104. Min Range: -0.8104

Max Act: 0.4897. Min Act: -0.0001

Data Index: 293526 (C4 (Web Text))

Max Activating Token Index: 893

Text #6

Max Range: 0.8104. Min Range: -0.8104

Max Act: 0.4975. Min Act: -0.0001

Data Index: 674496 (C4 (Web Text))

Max Activating Token Index: 519

Text #7

Max Range: 0.8104. Min Range: -0.8104

Max Act: 0.4949. Min Act: -0.0001

Data Index: 1064827 (C4 (Web Text))

Max Activating Token Index: 11

Text #8

Max Range: 0.8104. Min Range: -0.8104

Max Act: 0.4904. Min Act: -0.0001

Data Index: 911347 (C4 (Web Text))

Max Activating Token Index: 198

Text #9

Max Range: 0.8104. Min Range: -0.8104

Max Act: 0.5224. Min Act: -0.0001

Data Index: 1278940 (C4 (Web Text))

Max Activating Token Index: 927

Text #10

Max Range: 0.8104. Min Range: -0.8104

Max Act: 0.4780. Min Act: -0.0001

Data Index: 740377 (C4 (Web Text))

Max Activating Token Index: 173

Text #11

Max Range: 0.8104. Min Range: -0.8104

Max Act: 0.4779. Min Act: -0.0001

Data Index: 618864 (C4 (Web Text))

Max Activating Token Index: 353

Text #12

Max Range: 0.8104. Min Range: -0.8104

Max Act: 0.4457. Min Act: -0.0001

Data Index: 255821 (C4 (Web Text))

Max Activating Token Index: 442

Text #13

Max Range: 0.8104. Min Range: -0.8104

Max Act: 0.4489. Min Act: -0.0001

Data Index: 709742 (C4 (Web Text))

Max Activating Token Index: 46

Text #14

Max Range: 0.8104. Min Range: -0.8104

Max Act: 0.4725. Min Act: -0.0001

Data Index: 19605 (C4 (Web Text))

Max Activating Token Index: 942

Text #15

Max Range: 0.8104. Min Range: -0.8104

Max Act: 0.4386. Min Act: -0.0001

Data Index: 849073 (C4 (Web Text))

Max Activating Token Index: 538

Text #16

Max Range: 0.8104. Min Range: -0.8104

Max Act: 0.4723. Min Act: -0.0001

Data Index: 673607 (C4 (Web Text))

Max Activating Token Index: 230

Text #17

Max Range: 0.8104. Min Range: -0.8104

Max Act: 0.4521. Min Act: -0.0001

Data Index: 754422 (C4 (Web Text))

Max Activating Token Index: 103

Text #18

Max Range: 0.8104. Min Range: -0.8104

Max Act: 0.4687. Min Act: -0.0001

Data Index: 433089 (C4 (Web Text))

Max Activating Token Index: 549

Text #19

Max Range: 0.8104. Min Range: -0.8104

Max Act: 0.4432. Min Act: -0.0001

Data Index: 1156887 (C4 (Web Text))

Max Activating Token Index: 336