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

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

Neuron 749 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0072. Min Range: -0.0072

Max Act: 0.0072. Min Act: -0.0005

Data Index: 1366544 (Python Code)

Max Activating Token Index: 773

Text #1

Max Range: 0.0072. Min Range: -0.0072

Max Act: 0.0071. Min Act: -0.0005

Data Index: 1568235 (Python Code)

Max Activating Token Index: 1010

Text #2

Max Range: 0.0072. Min Range: -0.0072

Max Act: 0.0070. Min Act: -0.0006

Data Index: 199407 (C4 (Web Text))

Max Activating Token Index: 253

Text #3

Max Range: 0.0072. Min Range: -0.0072

Max Act: 0.0069. Min Act: -0.0005

Data Index: 1515717 (Python Code)

Max Activating Token Index: 734

Text #4

Max Range: 0.0072. Min Range: -0.0072

Max Act: 0.0069. Min Act: -0.0006

Data Index: 1164826 (C4 (Web Text))

Max Activating Token Index: 133

Text #5

Max Range: 0.0072. Min Range: -0.0072

Max Act: 0.0070. Min Act: -0.0005

Data Index: 1518463 (Python Code)

Max Activating Token Index: 384

Text #6

Max Range: 0.0072. Min Range: -0.0072

Max Act: 0.0070. Min Act: -0.0005

Data Index: 1530521 (Python Code)

Max Activating Token Index: 604

Text #7

Max Range: 0.0072. Min Range: -0.0072

Max Act: 0.0067. Min Act: -0.0006

Data Index: 1154224 (C4 (Web Text))

Max Activating Token Index: 461

Text #8

Max Range: 0.0072. Min Range: -0.0072

Max Act: 0.0069. Min Act: -0.0006

Data Index: 1267486 (C4 (Web Text))

Max Activating Token Index: 958

Text #9

Max Range: 0.0072. Min Range: -0.0072

Max Act: 0.0068. Min Act: -0.0006

Data Index: 690439 (C4 (Web Text))

Max Activating Token Index: 309

Text #10

Max Range: 0.0072. Min Range: -0.0072

Max Act: 0.0066. Min Act: -0.0006

Data Index: 329595 (C4 (Web Text))

Max Activating Token Index: 583

Text #11

Max Range: 0.0072. Min Range: -0.0072

Max Act: 0.0069. Min Act: -0.0005

Data Index: 1571959 (Python Code)

Max Activating Token Index: 734

Text #12

Max Range: 0.0072. Min Range: -0.0072

Max Act: 0.0068. Min Act: -0.0006

Data Index: 374816 (C4 (Web Text))

Max Activating Token Index: 584

Text #13

Max Range: 0.0072. Min Range: -0.0072

Max Act: 0.0068. Min Act: -0.0006

Data Index: 1522503 (Python Code)

Max Activating Token Index: 748

Text #14

Max Range: 0.0072. Min Range: -0.0072

Max Act: 0.0067. Min Act: -0.0005

Data Index: 1632081 (Python Code)

Max Activating Token Index: 736

Text #15

Max Range: 0.0072. Min Range: -0.0072

Max Act: 0.0067. Min Act: -0.0006

Data Index: 1321292 (C4 (Web Text))

Max Activating Token Index: 229

Text #16

Max Range: 0.0072. Min Range: -0.0072

Max Act: 0.0067. Min Act: -0.0006

Data Index: 1404562 (Python Code)

Max Activating Token Index: 696

Text #17

Max Range: 0.0072. Min Range: -0.0072

Max Act: 0.0067. Min Act: -0.0006

Data Index: 420508 (C4 (Web Text))

Max Activating Token Index: 498

Text #18

Max Range: 0.0072. Min Range: -0.0072

Max Act: 0.0067. Min Act: -0.0006

Data Index: 447183 (C4 (Web Text))

Max Activating Token Index: 669

Text #19

Max Range: 0.0072. Min Range: -0.0072

Max Act: 0.0066. Min Act: -0.0005

Data Index: 1501545 (Python Code)

Max Activating Token Index: 538