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

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

Neuron 617 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 1.1064. Min Range: -1.1064

Max Act: 1.1064. Min Act: -0.0001

Data Index: 260442 (C4 (Web Text))

Max Activating Token Index: 981

Text #1

Max Range: 1.1064. Min Range: -1.1064

Max Act: 1.0486. Min Act: -0.0001

Data Index: 367364 (C4 (Web Text))

Max Activating Token Index: 756

Text #2

Max Range: 1.1064. Min Range: -1.1064

Max Act: 0.9868. Min Act: -0.0001

Data Index: 784394 (C4 (Web Text))

Max Activating Token Index: 469

Text #3

Max Range: 1.1064. Min Range: -1.1064

Max Act: 0.9329. Min Act: -0.0001

Data Index: 1327650 (C4 (Web Text))

Max Activating Token Index: 235

Text #4

Max Range: 1.1064. Min Range: -1.1064

Max Act: 0.9189. Min Act: -0.0001

Data Index: 1333166 (C4 (Web Text))

Max Activating Token Index: 904

Text #5

Max Range: 1.1064. Min Range: -1.1064

Max Act: 0.9183. Min Act: -0.0001

Data Index: 1266123 (C4 (Web Text))

Max Activating Token Index: 155

Text #6

Max Range: 1.1064. Min Range: -1.1064

Max Act: 0.9152. Min Act: -0.0001

Data Index: 1270580 (C4 (Web Text))

Max Activating Token Index: 516

Text #7

Max Range: 1.1064. Min Range: -1.1064

Max Act: 0.9419. Min Act: -0.0001

Data Index: 790201 (C4 (Web Text))

Max Activating Token Index: 310

Text #8

Max Range: 1.1064. Min Range: -1.1064

Max Act: 0.8252. Min Act: -0.0001

Data Index: 367009 (C4 (Web Text))

Max Activating Token Index: 668

Text #9

Max Range: 1.1064. Min Range: -1.1064

Max Act: 0.8753. Min Act: -0.0001

Data Index: 239737 (C4 (Web Text))

Max Activating Token Index: 242

Text #10

Max Range: 1.1064. Min Range: -1.1064

Max Act: 0.8420. Min Act: -0.0001

Data Index: 312266 (C4 (Web Text))

Max Activating Token Index: 678

Text #11

Max Range: 1.1064. Min Range: -1.1064

Max Act: 0.8887. Min Act: -0.0001

Data Index: 311283 (C4 (Web Text))

Max Activating Token Index: 196

Text #12

Max Range: 1.1064. Min Range: -1.1064

Max Act: 0.8578. Min Act: -0.0001

Data Index: 173598 (C4 (Web Text))

Max Activating Token Index: 575

Text #13

Max Range: 1.1064. Min Range: -1.1064

Max Act: 0.8564. Min Act: -0.0001

Data Index: 1057504 (C4 (Web Text))

Max Activating Token Index: 783

Text #14

Max Range: 1.1064. Min Range: -1.1064

Max Act: 0.8534. Min Act: -0.0001

Data Index: 649804 (C4 (Web Text))

Max Activating Token Index: 42

Text #15

Max Range: 1.1064. Min Range: -1.1064

Max Act: 0.8848. Min Act: -0.0001

Data Index: 193398 (C4 (Web Text))

Max Activating Token Index: 771

Text #16

Max Range: 1.1064. Min Range: -1.1064

Max Act: 0.8577. Min Act: -0.0001

Data Index: 545242 (C4 (Web Text))

Max Activating Token Index: 275

Text #17

Max Range: 1.1064. Min Range: -1.1064

Max Act: 0.8563. Min Act: -0.0001

Data Index: 193991 (C4 (Web Text))

Max Activating Token Index: 884

Text #18

Max Range: 1.1064. Min Range: -1.1064

Max Act: 0.8362. Min Act: -0.0001

Data Index: 1248539 (C4 (Web Text))

Max Activating Token Index: 865

Text #19

Max Range: 1.1064. Min Range: -1.1064

Max Act: 0.8874. Min Act: -0.0001

Data Index: 295411 (C4 (Web Text))

Max Activating Token Index: 812