Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

Dataset: The Pile

Neuron 802 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l-pile')

Text #0

Max Range: 0.7700. Min Range: -0.7700

Max Act: 0.7700. Min Act: -0.0000

Data Index: 924236 (The Pile)

Max Activating Token Index: 94

Text #1

Max Range: 0.7700. Min Range: -0.7700

Max Act: 0.6907. Min Act: -0.0000

Data Index: 743257 (The Pile)

Max Activating Token Index: 283

Text #2

Max Range: 0.7700. Min Range: -0.7700

Max Act: 0.6228. Min Act: -0.0000

Data Index: 208024 (The Pile)

Max Activating Token Index: 678

Text #3

Max Range: 0.7700. Min Range: -0.7700

Max Act: 0.6047. Min Act: -0.0000

Data Index: 792625 (The Pile)

Max Activating Token Index: 455

Text #4

Max Range: 0.7700. Min Range: -0.7700

Max Act: 0.5947. Min Act: -0.0000

Data Index: 1783279 (The Pile)

Max Activating Token Index: 341

Text #5

Max Range: 0.7700. Min Range: -0.7700

Max Act: 0.5168. Min Act: -0.0000

Data Index: 1360301 (The Pile)

Max Activating Token Index: 438

Text #6

Max Range: 0.7700. Min Range: -0.7700

Max Act: 0.5158. Min Act: -0.0000

Data Index: 1626685 (The Pile)

Max Activating Token Index: 687

Text #7

Max Range: 0.7700. Min Range: -0.7700

Max Act: 0.4830. Min Act: -0.0000

Data Index: 1211255 (The Pile)

Max Activating Token Index: 589

Text #8

Max Range: 0.7700. Min Range: -0.7700

Max Act: 0.5233. Min Act: -0.0000

Data Index: 121131 (The Pile)

Max Activating Token Index: 91

Text #9

Max Range: 0.7700. Min Range: -0.7700

Max Act: 0.5026. Min Act: -0.0000

Data Index: 434690 (The Pile)

Max Activating Token Index: 118

Text #10

Max Range: 0.7700. Min Range: -0.7700

Max Act: 0.5377. Min Act: -0.0000

Data Index: 823490 (The Pile)

Max Activating Token Index: 107

Text #11

Max Range: 0.7700. Min Range: -0.7700

Max Act: 0.5483. Min Act: -0.0000

Data Index: 1996968 (The Pile)

Max Activating Token Index: 716

Text #12

Max Range: 0.7700. Min Range: -0.7700

Max Act: 0.5358. Min Act: -0.0000

Data Index: 1350398 (The Pile)

Max Activating Token Index: 821

Text #13

Max Range: 0.7700. Min Range: -0.7700

Max Act: 0.4769. Min Act: -0.0000

Data Index: 465612 (The Pile)

Max Activating Token Index: 175

Text #14

Max Range: 0.7700. Min Range: -0.7700

Max Act: 0.5238. Min Act: -0.0000

Data Index: 645361 (The Pile)

Max Activating Token Index: 241

Text #15

Max Range: 0.7700. Min Range: -0.7700

Max Act: 0.5085. Min Act: -0.0000

Data Index: 1464384 (The Pile)

Max Activating Token Index: 727

Text #16

Max Range: 0.7700. Min Range: -0.7700

Max Act: 0.4870. Min Act: -0.0000

Data Index: 1432107 (The Pile)

Max Activating Token Index: 324

Text #17

Max Range: 0.7700. Min Range: -0.7700

Max Act: 0.5025. Min Act: -0.0000

Data Index: 708262 (The Pile)

Max Activating Token Index: 617

Text #18

Max Range: 0.7700. Min Range: -0.7700

Max Act: 0.5444. Min Act: -0.0000

Data Index: 5236 (The Pile)

Max Activating Token Index: 371

Text #19

Max Range: 0.7700. Min Range: -0.7700

Max Act: 0.5106. Min Act: -0.0000

Data Index: 1473255 (The Pile)

Max Activating Token Index: 482