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

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

Neuron 3768 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.4073. Min Range: -0.4073

Max Act: 0.4040. Min Act: -0.0000

Data Index: 849517 (C4 (Web Text))

Max Activating Token Index: 427

Text #1

Max Range: 0.4073. Min Range: -0.4073

Max Act: 0.3859. Min Act: -0.0000

Data Index: 267794 (C4 (Web Text))

Max Activating Token Index: 27

Text #2

Max Range: 0.4073. Min Range: -0.4073

Max Act: 0.4073. Min Act: -0.0000

Data Index: 296132 (C4 (Web Text))

Max Activating Token Index: 131

Text #3

Max Range: 0.4073. Min Range: -0.4073

Max Act: 0.3784. Min Act: -0.0000

Data Index: 524046 (C4 (Web Text))

Max Activating Token Index: 413

Text #4

Max Range: 0.4073. Min Range: -0.4073

Max Act: 0.3856. Min Act: -0.0000

Data Index: 675389 (C4 (Web Text))

Max Activating Token Index: 463

Text #5

Max Range: 0.4073. Min Range: -0.4073

Max Act: 0.3686. Min Act: -0.0000

Data Index: 100795 (C4 (Web Text))

Max Activating Token Index: 397

Text #6

Max Range: 0.4073. Min Range: -0.4073

Max Act: 0.4000. Min Act: -0.0000

Data Index: 845401 (C4 (Web Text))

Max Activating Token Index: 222

Text #7

Max Range: 0.4073. Min Range: -0.4073

Max Act: 0.3719. Min Act: -0.0000

Data Index: 1084261 (C4 (Web Text))

Max Activating Token Index: 525

Text #8

Max Range: 0.4073. Min Range: -0.4073

Max Act: 0.3682. Min Act: -0.0000

Data Index: 887218 (C4 (Web Text))

Max Activating Token Index: 876

Text #9

Max Range: 0.4073. Min Range: -0.4073

Max Act: 0.3550. Min Act: -0.0000

Data Index: 759507 (C4 (Web Text))

Max Activating Token Index: 651

Text #10

Max Range: 0.4073. Min Range: -0.4073

Max Act: 0.3301. Min Act: -0.0000

Data Index: 1104982 (C4 (Web Text))

Max Activating Token Index: 847

Text #11

Max Range: 0.4073. Min Range: -0.4073

Max Act: 0.3241. Min Act: -0.0000

Data Index: 1082171 (C4 (Web Text))

Max Activating Token Index: 91

Text #12

Max Range: 0.4073. Min Range: -0.4073

Max Act: 0.3124. Min Act: -0.0000

Data Index: 457027 (C4 (Web Text))

Max Activating Token Index: 926

Text #13

Max Range: 0.4073. Min Range: -0.4073

Max Act: 0.3315. Min Act: -0.0000

Data Index: 705091 (C4 (Web Text))

Max Activating Token Index: 197

Text #14

Max Range: 0.4073. Min Range: -0.4073

Max Act: 0.2962. Min Act: -0.0000

Data Index: 891836 (C4 (Web Text))

Max Activating Token Index: 258

Text #15

Max Range: 0.4073. Min Range: -0.4073

Max Act: 0.2980. Min Act: -0.0000

Data Index: 804972 (C4 (Web Text))

Max Activating Token Index: 204

Text #16

Max Range: 0.4073. Min Range: -0.4073

Max Act: 0.3064. Min Act: -0.0000

Data Index: 1105001 (C4 (Web Text))

Max Activating Token Index: 150

Text #17

Max Range: 0.4073. Min Range: -0.4073

Max Act: 0.3145. Min Act: -0.0000

Data Index: 12345 (C4 (Web Text))

Max Activating Token Index: 845

Text #18

Max Range: 0.4073. Min Range: -0.4073

Max Act: 0.2869. Min Act: -0.0000

Data Index: 608528 (C4 (Web Text))

Max Activating Token Index: 159

Text #19

Max Range: 0.4073. Min Range: -0.4073

Max Act: 0.3047. Min Act: -0.0000

Data Index: 53007 (C4 (Web Text))

Max Activating Token Index: 24