Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 683 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.2560. Min Range: -0.2560

Max Act: 0.2560. Min Act: -0.0001

Data Index: 1204302 (C4 (Web Text))

Max Activating Token Index: 819

Text #1

Max Range: 0.2560. Min Range: -0.2560

Max Act: 0.2435. Min Act: -0.0001

Data Index: 1178648 (C4 (Web Text))

Max Activating Token Index: 943

Text #2

Max Range: 0.2560. Min Range: -0.2560

Max Act: 0.2380. Min Act: -0.0001

Data Index: 841905 (C4 (Web Text))

Max Activating Token Index: 236

Text #3

Max Range: 0.2560. Min Range: -0.2560

Max Act: 0.2268. Min Act: -0.0001

Data Index: 947354 (C4 (Web Text))

Max Activating Token Index: 55

Text #4

Max Range: 0.2560. Min Range: -0.2560

Max Act: 0.2234. Min Act: -0.0001

Data Index: 902517 (C4 (Web Text))

Max Activating Token Index: 223

Text #5

Max Range: 0.2560. Min Range: -0.2560

Max Act: 0.2296. Min Act: -0.0001

Data Index: 460683 (C4 (Web Text))

Max Activating Token Index: 647

Text #6

Max Range: 0.2560. Min Range: -0.2560

Max Act: 0.2132. Min Act: -0.0001

Data Index: 391395 (C4 (Web Text))

Max Activating Token Index: 171

Text #7

Max Range: 0.2560. Min Range: -0.2560

Max Act: 0.2056. Min Act: -0.0001

Data Index: 557266 (C4 (Web Text))

Max Activating Token Index: 124

Text #8

Max Range: 0.2560. Min Range: -0.2560

Max Act: 0.2065. Min Act: -0.0001

Data Index: 1035088 (C4 (Web Text))

Max Activating Token Index: 1005

Text #9

Max Range: 0.2560. Min Range: -0.2560

Max Act: 0.2101. Min Act: -0.0001

Data Index: 599359 (C4 (Web Text))

Max Activating Token Index: 141

Text #10

Max Range: 0.2560. Min Range: -0.2560

Max Act: 0.2004. Min Act: -0.0001

Data Index: 622971 (C4 (Web Text))

Max Activating Token Index: 152

Text #11

Max Range: 0.2560. Min Range: -0.2560

Max Act: 0.2038. Min Act: -0.0001

Data Index: 1078746 (C4 (Web Text))

Max Activating Token Index: 11

Text #12

Max Range: 0.2560. Min Range: -0.2560

Max Act: 0.1960. Min Act: -0.0001

Data Index: 1208727 (C4 (Web Text))

Max Activating Token Index: 217

Text #13

Max Range: 0.2560. Min Range: -0.2560

Max Act: 0.2019. Min Act: -0.0001

Data Index: 567713 (C4 (Web Text))

Max Activating Token Index: 746

Text #14

Max Range: 0.2560. Min Range: -0.2560

Max Act: 0.2034. Min Act: -0.0001

Data Index: 168998 (C4 (Web Text))

Max Activating Token Index: 299

Text #15

Max Range: 0.2560. Min Range: -0.2560

Max Act: 0.2016. Min Act: -0.0001

Data Index: 686889 (C4 (Web Text))

Max Activating Token Index: 544

Text #16

Max Range: 0.2560. Min Range: -0.2560

Max Act: 0.2026. Min Act: -0.0001

Data Index: 514075 (C4 (Web Text))

Max Activating Token Index: 421

Text #17

Max Range: 0.2560. Min Range: -0.2560

Max Act: 0.1939. Min Act: -0.0001

Data Index: 654562 (C4 (Web Text))

Max Activating Token Index: 29

Text #18

Max Range: 0.2560. Min Range: -0.2560

Max Act: 0.1943. Min Act: -0.0001

Data Index: 621393 (C4 (Web Text))

Max Activating Token Index: 343

Text #19

Max Range: 0.2560. Min Range: -0.2560

Max Act: 0.1943. Min Act: -0.0001

Data Index: 1355768 (C4 (Web Text))

Max Activating Token Index: 366