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

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

Neuron 3839 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7103. Min Range: -0.7103

Max Act: 0.6209. Min Act: -0.0001

Data Index: 810958 (C4 (Web Text))

Max Activating Token Index: 746

Text #1

Max Range: 0.7103. Min Range: -0.7103

Max Act: 0.5915. Min Act: -0.0001

Data Index: 1368311 (Python Code)

Max Activating Token Index: 791

Text #2

Max Range: 0.7103. Min Range: -0.7103

Max Act: 0.6841. Min Act: -0.0001

Data Index: 192360 (C4 (Web Text))

Max Activating Token Index: 118

Text #3

Max Range: 0.7103. Min Range: -0.7103

Max Act: 0.5482. Min Act: -0.0001

Data Index: 1377218 (Python Code)

Max Activating Token Index: 795

Text #4

Max Range: 0.7103. Min Range: -0.7103

Max Act: 0.7103. Min Act: -0.0001

Data Index: 435104 (C4 (Web Text))

Max Activating Token Index: 786

Text #5

Max Range: 0.7103. Min Range: -0.7103

Max Act: 0.5669. Min Act: -0.0001

Data Index: 1339540 (C4 (Web Text))

Max Activating Token Index: 663

Text #6

Max Range: 0.7103. Min Range: -0.7103

Max Act: 0.6423. Min Act: -0.0001

Data Index: 227506 (C4 (Web Text))

Max Activating Token Index: 503

Text #7

Max Range: 0.7103. Min Range: -0.7103

Max Act: 0.5598. Min Act: -0.0001

Data Index: 165389 (C4 (Web Text))

Max Activating Token Index: 941

Text #8

Max Range: 0.7103. Min Range: -0.7103

Max Act: 0.5776. Min Act: -0.0001

Data Index: 204218 (C4 (Web Text))

Max Activating Token Index: 137

Text #9

Max Range: 0.7103. Min Range: -0.7103

Max Act: 0.5901. Min Act: -0.0001

Data Index: 1054631 (C4 (Web Text))

Max Activating Token Index: 130

Text #10

Max Range: 0.7103. Min Range: -0.7103

Max Act: 0.5741. Min Act: -0.0001

Data Index: 1214077 (C4 (Web Text))

Max Activating Token Index: 174

Text #11

Max Range: 0.7103. Min Range: -0.7103

Max Act: 0.4809. Min Act: -0.0001

Data Index: 260116 (C4 (Web Text))

Max Activating Token Index: 870

Text #12

Max Range: 0.7103. Min Range: -0.7103

Max Act: 0.5784. Min Act: -0.0001

Data Index: 1392017 (Python Code)

Max Activating Token Index: 827

Text #13

Max Range: 0.7103. Min Range: -0.7103

Max Act: 0.4609. Min Act: -0.0001

Data Index: 1535958 (Python Code)

Max Activating Token Index: 231

Text #14

Max Range: 0.7103. Min Range: -0.7103

Max Act: 0.5511. Min Act: -0.0001

Data Index: 107054 (C4 (Web Text))

Max Activating Token Index: 561

Text #15

Max Range: 0.7103. Min Range: -0.7103

Max Act: 0.5531. Min Act: -0.0001

Data Index: 249229 (C4 (Web Text))

Max Activating Token Index: 956

Text #16

Max Range: 0.7103. Min Range: -0.7103

Max Act: 0.4939. Min Act: -0.0001

Data Index: 76078 (C4 (Web Text))

Max Activating Token Index: 803

Text #17

Max Range: 0.7103. Min Range: -0.7103

Max Act: 0.5218. Min Act: -0.0001

Data Index: 100929 (C4 (Web Text))

Max Activating Token Index: 77

Text #18

Max Range: 0.7103. Min Range: -0.7103

Max Act: 0.5236. Min Act: -0.0001

Data Index: 68938 (C4 (Web Text))

Max Activating Token Index: 276

Text #19

Max Range: 0.7103. Min Range: -0.7103

Max Act: 0.6014. Min Act: -0.0001

Data Index: 137945 (C4 (Web Text))

Max Activating Token Index: 371