Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1263 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 8.4944. Min Range: -8.4944

Max Act: 8.3069. Min Act: -0.1700

Data Index: 1150958 (C4 (Web Text))

Max Activating Token Index: 1011

Text #1

Max Range: 8.4944. Min Range: -8.4944

Max Act: 8.4944. Min Act: -0.1700

Data Index: 978116 (C4 (Web Text))

Max Activating Token Index: 399

Text #2

Max Range: 8.4944. Min Range: -8.4944

Max Act: 8.3069. Min Act: -0.1700

Data Index: 1328840 (C4 (Web Text))

Max Activating Token Index: 796

Text #3

Max Range: 8.4944. Min Range: -8.4944

Max Act: 8.1819. Min Act: -0.1698

Data Index: 980755 (C4 (Web Text))

Max Activating Token Index: 248

Text #4

Max Range: 8.4944. Min Range: -8.4944

Max Act: 8.4319. Min Act: -0.1700

Data Index: 157921 (C4 (Web Text))

Max Activating Token Index: 583

Text #5

Max Range: 8.4944. Min Range: -8.4944

Max Act: 8.4319. Min Act: -0.1700

Data Index: 1290151 (C4 (Web Text))

Max Activating Token Index: 379

Text #6

Max Range: 8.4944. Min Range: -8.4944

Max Act: 8.3069. Min Act: -0.1700

Data Index: 341931 (C4 (Web Text))

Max Activating Token Index: 307

Text #7

Max Range: 8.4944. Min Range: -8.4944

Max Act: 8.3069. Min Act: -0.1700

Data Index: 205839 (C4 (Web Text))

Max Activating Token Index: 285

Text #8

Max Range: 8.4944. Min Range: -8.4944

Max Act: 8.1194. Min Act: -0.1700

Data Index: 1046336 (C4 (Web Text))

Max Activating Token Index: 715

Text #9

Max Range: 8.4944. Min Range: -8.4944

Max Act: 8.3069. Min Act: -0.1700

Data Index: 338664 (C4 (Web Text))

Max Activating Token Index: 565

Text #10

Max Range: 8.4944. Min Range: -8.4944

Max Act: 8.1819. Min Act: -0.1700

Data Index: 861523 (C4 (Web Text))

Max Activating Token Index: 225

Text #11

Max Range: 8.4944. Min Range: -8.4944

Max Act: 8.3069. Min Act: -0.1700

Data Index: 615099 (C4 (Web Text))

Max Activating Token Index: 808

Text #12

Max Range: 8.4944. Min Range: -8.4944

Max Act: 8.1819. Min Act: -0.1700

Data Index: 168814 (C4 (Web Text))

Max Activating Token Index: 304

Text #13

Max Range: 8.4944. Min Range: -8.4944

Max Act: 7.9944. Min Act: -0.1700

Data Index: 251969 (C4 (Web Text))

Max Activating Token Index: 12

Text #14

Max Range: 8.4944. Min Range: -8.4944

Max Act: 8.0569. Min Act: -0.1700

Data Index: 170421 (C4 (Web Text))

Max Activating Token Index: 556

Text #15

Max Range: 8.4944. Min Range: -8.4944

Max Act: 8.2444. Min Act: -0.1700

Data Index: 140147 (C4 (Web Text))

Max Activating Token Index: 140

Text #16

Max Range: 8.4944. Min Range: -8.4944

Max Act: 8.1819. Min Act: -0.1700

Data Index: 501355 (C4 (Web Text))

Max Activating Token Index: 598

Text #17

Max Range: 8.4944. Min Range: -8.4944

Max Act: 8.3069. Min Act: -0.1700

Data Index: 790436 (C4 (Web Text))

Max Activating Token Index: 128

Text #18

Max Range: 8.4944. Min Range: -8.4944

Max Act: 7.9944. Min Act: -0.1700

Data Index: 534512 (C4 (Web Text))

Max Activating Token Index: 449

Text #19

Max Range: 8.4944. Min Range: -8.4944

Max Act: 8.1194. Min Act: -0.1700

Data Index: 180931 (C4 (Web Text))

Max Activating Token Index: 743