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

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

Neuron 2020 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Click the toggle to see the full text!

Text #0

Max Range: 0.0403. Min Range: -0.0403

Max Act: 0.0403. Min Act: -0.0001

Data Index: 374512 (C4 (Web Text))

Max Activating Token Index: 40

Text #1

Max Range: 0.0403. Min Range: -0.0403

Max Act: 0.0401. Min Act: -0.0001

Data Index: 271507 (C4 (Web Text))

Max Activating Token Index: 400

Text #2

Max Range: 0.0403. Min Range: -0.0403

Max Act: 0.0387. Min Act: -0.0001

Data Index: 1345161 (C4 (Web Text))

Max Activating Token Index: 115

Text #3

Max Range: 0.0403. Min Range: -0.0403

Max Act: 0.0387. Min Act: -0.0001

Data Index: 1371442 (Python Code)

Max Activating Token Index: 89

Text #4

Max Range: 0.0403. Min Range: -0.0403

Max Act: 0.0384. Min Act: -0.0001

Data Index: 1487048 (Python Code)

Max Activating Token Index: 85

Text #5

Max Range: 0.0403. Min Range: -0.0403

Max Act: 0.0372. Min Act: -0.0001

Data Index: 969920 (C4 (Web Text))

Max Activating Token Index: 778

Text #6

Max Range: 0.0403. Min Range: -0.0403

Max Act: 0.0370. Min Act: -0.0001

Data Index: 1229839 (C4 (Web Text))

Max Activating Token Index: 82

Text #7

Max Range: 0.0403. Min Range: -0.0403

Max Act: 0.0370. Min Act: -0.0001

Data Index: 1171993 (C4 (Web Text))

Max Activating Token Index: 262

Text #8

Max Range: 0.0403. Min Range: -0.0403

Max Act: 0.0374. Min Act: -0.0001

Data Index: 1491642 (Python Code)

Max Activating Token Index: 168

Text #9

Max Range: 0.0403. Min Range: -0.0403

Max Act: 0.0371. Min Act: -0.0001

Data Index: 1630746 (Python Code)

Max Activating Token Index: 10

Text #10

Max Range: 0.0403. Min Range: -0.0403

Max Act: 0.0367. Min Act: -0.0001

Data Index: 1632678 (Python Code)

Max Activating Token Index: 36

Text #11

Max Range: 0.0403. Min Range: -0.0403

Max Act: 0.0369. Min Act: -0.0002

Data Index: 520979 (C4 (Web Text))

Max Activating Token Index: 89

Text #12

Max Range: 0.0403. Min Range: -0.0403

Max Act: 0.0365. Min Act: -0.0001

Data Index: 1175383 (C4 (Web Text))

Max Activating Token Index: 129

Text #13

Max Range: 0.0403. Min Range: -0.0403

Max Act: 0.0376. Min Act: -0.0001

Data Index: 1453275 (Python Code)

Max Activating Token Index: 348

Text #14

Max Range: 0.0403. Min Range: -0.0403

Max Act: 0.0367. Min Act: -0.0001

Data Index: 372439 (C4 (Web Text))

Max Activating Token Index: 447

Text #15

Max Range: 0.0403. Min Range: -0.0403

Max Act: 0.0367. Min Act: -0.0001

Data Index: 724775 (C4 (Web Text))

Max Activating Token Index: 118

Text #16

Max Range: 0.0403. Min Range: -0.0403

Max Act: 0.0360. Min Act: -0.0001

Data Index: 1321824 (C4 (Web Text))

Max Activating Token Index: 86

Text #17

Max Range: 0.0403. Min Range: -0.0403

Max Act: 0.0364. Min Act: -0.0001

Data Index: 91261 (C4 (Web Text))

Max Activating Token Index: 366

Text #18

Max Range: 0.0403. Min Range: -0.0403

Max Act: 0.0356. Min Act: -0.0001

Data Index: 1211452 (C4 (Web Text))

Max Activating Token Index: 22

Text #19

Max Range: 0.0403. Min Range: -0.0403

Max Act: 0.0363. Min Act: -0.0001

Data Index: 322288 (C4 (Web Text))

Max Activating Token Index: 25