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

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

Neuron 4034 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4494. Min Range: -0.4494

Max Act: 0.4494. Min Act: -0.0001

Data Index: 780317 (C4 (Web Text))

Max Activating Token Index: 882

Text #1

Max Range: 0.4494. Min Range: -0.4494

Max Act: 0.3686. Min Act: -0.0001

Data Index: 253599 (C4 (Web Text))

Max Activating Token Index: 210

Text #2

Max Range: 0.4494. Min Range: -0.4494

Max Act: 0.3312. Min Act: -0.0001

Data Index: 7677 (C4 (Web Text))

Max Activating Token Index: 524

Text #3

Max Range: 0.4494. Min Range: -0.4494

Max Act: 0.3017. Min Act: -0.0001

Data Index: 259850 (C4 (Web Text))

Max Activating Token Index: 552

Text #4

Max Range: 0.4494. Min Range: -0.4494

Max Act: 0.2956. Min Act: -0.0001

Data Index: 138928 (C4 (Web Text))

Max Activating Token Index: 284

Text #5

Max Range: 0.4494. Min Range: -0.4494

Max Act: 0.3075. Min Act: -0.0001

Data Index: 527149 (C4 (Web Text))

Max Activating Token Index: 484

Text #6

Max Range: 0.4494. Min Range: -0.4494

Max Act: 0.2959. Min Act: -0.0001

Data Index: 809613 (C4 (Web Text))

Max Activating Token Index: 414

Text #7

Max Range: 0.4494. Min Range: -0.4494

Max Act: 0.3055. Min Act: -0.0001

Data Index: 776531 (C4 (Web Text))

Max Activating Token Index: 94

Text #8

Max Range: 0.4494. Min Range: -0.4494

Max Act: 0.3140. Min Act: -0.0001

Data Index: 48642 (C4 (Web Text))

Max Activating Token Index: 824

Text #9

Max Range: 0.4494. Min Range: -0.4494

Max Act: 0.3014. Min Act: -0.0001

Data Index: 970303 (C4 (Web Text))

Max Activating Token Index: 567

Text #10

Max Range: 0.4494. Min Range: -0.4494

Max Act: 0.3108. Min Act: -0.0001

Data Index: 888338 (C4 (Web Text))

Max Activating Token Index: 883

Text #11

Max Range: 0.4494. Min Range: -0.4494

Max Act: 0.2826. Min Act: -0.0001

Data Index: 1267556 (C4 (Web Text))

Max Activating Token Index: 236

Text #12

Max Range: 0.4494. Min Range: -0.4494

Max Act: 0.3026. Min Act: -0.0001

Data Index: 10412 (C4 (Web Text))

Max Activating Token Index: 105

Text #13

Max Range: 0.4494. Min Range: -0.4494

Max Act: 0.2898. Min Act: -0.0001

Data Index: 336055 (C4 (Web Text))

Max Activating Token Index: 915

Text #14

Max Range: 0.4494. Min Range: -0.4494

Max Act: 0.2651. Min Act: -0.0001

Data Index: 128086 (C4 (Web Text))

Max Activating Token Index: 705

Text #15

Max Range: 0.4494. Min Range: -0.4494

Max Act: 0.2687. Min Act: -0.0001

Data Index: 1035694 (C4 (Web Text))

Max Activating Token Index: 72

Text #16

Max Range: 0.4494. Min Range: -0.4494

Max Act: 0.2819. Min Act: -0.0001

Data Index: 1396305 (Python Code)

Max Activating Token Index: 193

Text #17

Max Range: 0.4494. Min Range: -0.4494

Max Act: 0.2863. Min Act: -0.0001

Data Index: 734049 (C4 (Web Text))

Max Activating Token Index: 384

Text #18

Max Range: 0.4494. Min Range: -0.4494

Max Act: 0.2825. Min Act: -0.0001

Data Index: 1132772 (C4 (Web Text))

Max Activating Token Index: 1005

Text #19

Max Range: 0.4494. Min Range: -0.4494

Max Act: 0.2623. Min Act: -0.0001

Data Index: 1346585 (C4 (Web Text))

Max Activating Token Index: 724