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

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

Neuron 676 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4280. Min Range: -0.4280

Max Act: 0.4280. Min Act: -0.0001

Data Index: 1272857 (C4 (Web Text))

Max Activating Token Index: 527

Text #1

Max Range: 0.4280. Min Range: -0.4280

Max Act: 0.3818. Min Act: -0.0001

Data Index: 333316 (C4 (Web Text))

Max Activating Token Index: 442

Text #2

Max Range: 0.4280. Min Range: -0.4280

Max Act: 0.3215. Min Act: -0.0000

Data Index: 1395295 (Python Code)

Max Activating Token Index: 604

Text #3

Max Range: 0.4280. Min Range: -0.4280

Max Act: 0.3037. Min Act: -0.0001

Data Index: 463375 (C4 (Web Text))

Max Activating Token Index: 164

Text #4

Max Range: 0.4280. Min Range: -0.4280

Max Act: 0.2858. Min Act: -0.0000

Data Index: 1440877 (Python Code)

Max Activating Token Index: 193

Text #5

Max Range: 0.4280. Min Range: -0.4280

Max Act: 0.2956. Min Act: -0.0001

Data Index: 919711 (C4 (Web Text))

Max Activating Token Index: 152

Text #6

Max Range: 0.4280. Min Range: -0.4280

Max Act: 0.3028. Min Act: -0.0001

Data Index: 1059294 (C4 (Web Text))

Max Activating Token Index: 31

Text #7

Max Range: 0.4280. Min Range: -0.4280

Max Act: 0.3042. Min Act: -0.0001

Data Index: 1515629 (Python Code)

Max Activating Token Index: 438

Text #8

Max Range: 0.4280. Min Range: -0.4280

Max Act: 0.3069. Min Act: -0.0001

Data Index: 444791 (C4 (Web Text))

Max Activating Token Index: 163

Text #9

Max Range: 0.4280. Min Range: -0.4280

Max Act: 0.2896. Min Act: -0.0001

Data Index: 1364291 (Python Code)

Max Activating Token Index: 341

Text #10

Max Range: 0.4280. Min Range: -0.4280

Max Act: 0.2976. Min Act: -0.0001

Data Index: 423657 (C4 (Web Text))

Max Activating Token Index: 863

Text #11

Max Range: 0.4280. Min Range: -0.4280

Max Act: 0.3136. Min Act: -0.0001

Data Index: 1001781 (C4 (Web Text))

Max Activating Token Index: 319

Text #12

Max Range: 0.4280. Min Range: -0.4280

Max Act: 0.2916. Min Act: -0.0001

Data Index: 1013359 (C4 (Web Text))

Max Activating Token Index: 494

Text #13

Max Range: 0.4280. Min Range: -0.4280

Max Act: 0.2939. Min Act: -0.0001

Data Index: 1056932 (C4 (Web Text))

Max Activating Token Index: 167

Text #14

Max Range: 0.4280. Min Range: -0.4280

Max Act: 0.2786. Min Act: -0.0001

Data Index: 1276955 (C4 (Web Text))

Max Activating Token Index: 262

Text #15

Max Range: 0.4280. Min Range: -0.4280

Max Act: 0.2799. Min Act: -0.0001

Data Index: 911034 (C4 (Web Text))

Max Activating Token Index: 896

Text #16

Max Range: 0.4280. Min Range: -0.4280

Max Act: 0.2725. Min Act: -0.0001

Data Index: 109687 (C4 (Web Text))

Max Activating Token Index: 98

Text #17

Max Range: 0.4280. Min Range: -0.4280

Max Act: 0.2812. Min Act: -0.0001

Data Index: 1415046 (Python Code)

Max Activating Token Index: 291

Text #18

Max Range: 0.4280. Min Range: -0.4280

Max Act: 0.2817. Min Act: -0.0001

Data Index: 1239085 (C4 (Web Text))

Max Activating Token Index: 223

Text #19

Max Range: 0.4280. Min Range: -0.4280

Max Act: 0.2831. Min Act: -0.0000

Data Index: 1468321 (Python Code)

Max Activating Token Index: 160