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

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

Neuron 2620 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5374. Min Range: -0.5374

Max Act: 0.5307. Min Act: -0.0001

Data Index: 656304 (C4 (Web Text))

Max Activating Token Index: 290

Text #1

Max Range: 0.5374. Min Range: -0.5374

Max Act: 0.5073. Min Act: -0.0001

Data Index: 168564 (C4 (Web Text))

Max Activating Token Index: 1009

Text #2

Max Range: 0.5374. Min Range: -0.5374

Max Act: 0.5374. Min Act: -0.0001

Data Index: 893639 (C4 (Web Text))

Max Activating Token Index: 958

Text #3

Max Range: 0.5374. Min Range: -0.5374

Max Act: 0.5169. Min Act: -0.0001

Data Index: 1504777 (Python Code)

Max Activating Token Index: 927

Text #4

Max Range: 0.5374. Min Range: -0.5374

Max Act: 0.4951. Min Act: -0.0001

Data Index: 961507 (C4 (Web Text))

Max Activating Token Index: 533

Text #5

Max Range: 0.5374. Min Range: -0.5374

Max Act: 0.4714. Min Act: -0.0001

Data Index: 720755 (C4 (Web Text))

Max Activating Token Index: 187

Text #6

Max Range: 0.5374. Min Range: -0.5374

Max Act: 0.4524. Min Act: -0.0001

Data Index: 1234646 (C4 (Web Text))

Max Activating Token Index: 866

Text #7

Max Range: 0.5374. Min Range: -0.5374

Max Act: 0.4699. Min Act: -0.0001

Data Index: 1380156 (Python Code)

Max Activating Token Index: 921

Text #8

Max Range: 0.5374. Min Range: -0.5374

Max Act: 0.4835. Min Act: -0.0001

Data Index: 806594 (C4 (Web Text))

Max Activating Token Index: 1007

Text #9

Max Range: 0.5374. Min Range: -0.5374

Max Act: 0.4641. Min Act: -0.0001

Data Index: 274219 (C4 (Web Text))

Max Activating Token Index: 482

Text #10

Max Range: 0.5374. Min Range: -0.5374

Max Act: 0.4422. Min Act: -0.0001

Data Index: 463694 (C4 (Web Text))

Max Activating Token Index: 97

Text #11

Max Range: 0.5374. Min Range: -0.5374

Max Act: 0.4429. Min Act: -0.0001

Data Index: 783888 (C4 (Web Text))

Max Activating Token Index: 181

Text #12

Max Range: 0.5374. Min Range: -0.5374

Max Act: 0.4135. Min Act: -0.0001

Data Index: 1124246 (C4 (Web Text))

Max Activating Token Index: 588

Text #13

Max Range: 0.5374. Min Range: -0.5374

Max Act: 0.4548. Min Act: -0.0001

Data Index: 384069 (C4 (Web Text))

Max Activating Token Index: 621

Text #14

Max Range: 0.5374. Min Range: -0.5374

Max Act: 0.4514. Min Act: -0.0001

Data Index: 1075142 (C4 (Web Text))

Max Activating Token Index: 947

Text #15

Max Range: 0.5374. Min Range: -0.5374

Max Act: 0.4664. Min Act: -0.0001

Data Index: 530988 (C4 (Web Text))

Max Activating Token Index: 486

Text #16

Max Range: 0.5374. Min Range: -0.5374

Max Act: 0.4451. Min Act: -0.0001

Data Index: 1222604 (C4 (Web Text))

Max Activating Token Index: 849

Text #17

Max Range: 0.5374. Min Range: -0.5374

Max Act: 0.4438. Min Act: -0.0001

Data Index: 1361955 (Python Code)

Max Activating Token Index: 724

Text #18

Max Range: 0.5374. Min Range: -0.5374

Max Act: 0.4578. Min Act: -0.0001

Data Index: 1106274 (C4 (Web Text))

Max Activating Token Index: 232

Text #19

Max Range: 0.5374. Min Range: -0.5374

Max Act: 0.4250. Min Act: -0.0001

Data Index: 870563 (C4 (Web Text))

Max Activating Token Index: 618