Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 620 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 2.9385. Min Range: -2.9385

Max Act: 2.8819. Min Act: -0.0000

Data Index: 1553860 (Python Code)

Max Activating Token Index: 784

Text #1

Max Range: 2.9385. Min Range: -2.9385

Max Act: 2.8311. Min Act: -0.0000

Data Index: 1505938 (Python Code)

Max Activating Token Index: 277

Text #2

Max Range: 2.9385. Min Range: -2.9385

Max Act: 2.6744. Min Act: -0.0000

Data Index: 511198 (C4 (Web Text))

Max Activating Token Index: 76

Text #3

Max Range: 2.9385. Min Range: -2.9385

Max Act: 2.9385. Min Act: -0.0000

Data Index: 350473 (C4 (Web Text))

Max Activating Token Index: 98

Text #4

Max Range: 2.9385. Min Range: -2.9385

Max Act: 2.7644. Min Act: -0.0000

Data Index: 892538 (C4 (Web Text))

Max Activating Token Index: 592

Text #5

Max Range: 2.9385. Min Range: -2.9385

Max Act: 2.7620. Min Act: -0.0000

Data Index: 842143 (C4 (Web Text))

Max Activating Token Index: 542

Text #6

Max Range: 2.9385. Min Range: -2.9385

Max Act: 2.6001. Min Act: -0.0000

Data Index: 1175716 (C4 (Web Text))

Max Activating Token Index: 488

Text #7

Max Range: 2.9385. Min Range: -2.9385

Max Act: 2.3366. Min Act: -0.0000

Data Index: 626931 (C4 (Web Text))

Max Activating Token Index: 899

Text #8

Max Range: 2.9385. Min Range: -2.9385

Max Act: 2.5301. Min Act: -0.0000

Data Index: 561850 (C4 (Web Text))

Max Activating Token Index: 141

Text #9

Max Range: 2.9385. Min Range: -2.9385

Max Act: 2.6205. Min Act: -0.0000

Data Index: 429545 (C4 (Web Text))

Max Activating Token Index: 841

Text #10

Max Range: 2.9385. Min Range: -2.9385

Max Act: 2.3220. Min Act: -0.0000

Data Index: 1360802 (Python Code)

Max Activating Token Index: 99

Text #11

Max Range: 2.9385. Min Range: -2.9385

Max Act: 2.2894. Min Act: -0.0000

Data Index: 1589055 (Python Code)

Max Activating Token Index: 285

Text #12

Max Range: 2.9385. Min Range: -2.9385

Max Act: 2.1087. Min Act: -0.0000

Data Index: 546081 (C4 (Web Text))

Max Activating Token Index: 885

Text #13

Max Range: 2.9385. Min Range: -2.9385

Max Act: 2.3200. Min Act: -0.0000

Data Index: 1488393 (Python Code)

Max Activating Token Index: 441

Text #14

Max Range: 2.9385. Min Range: -2.9385

Max Act: 2.4400. Min Act: -0.0000

Data Index: 1597738 (Python Code)

Max Activating Token Index: 241

Text #15

Max Range: 2.9385. Min Range: -2.9385

Max Act: 2.1932. Min Act: -0.0000

Data Index: 1554546 (Python Code)

Max Activating Token Index: 53

Text #16

Max Range: 2.9385. Min Range: -2.9385

Max Act: 2.1912. Min Act: -0.0000

Data Index: 1234618 (C4 (Web Text))

Max Activating Token Index: 349

Text #17

Max Range: 2.9385. Min Range: -2.9385

Max Act: 2.2669. Min Act: -0.0000

Data Index: 376187 (C4 (Web Text))

Max Activating Token Index: 651

Text #18

Max Range: 2.9385. Min Range: -2.9385

Max Act: 2.2467. Min Act: -0.0000

Data Index: 947736 (C4 (Web Text))

Max Activating Token Index: 100

Text #19

Max Range: 2.9385. Min Range: -2.9385

Max Act: 2.2279. Min Act: -0.0000

Data Index: 1415558 (Python Code)

Max Activating Token Index: 635