Model: SoLU Model: 8 Layers, 4096 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 28 in Layer 4
Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')
Text #0
Max Range: 1.4581. Min Range: -1.4581
Max Act: 1.4581. Min Act: -0.0001
Data Index: 910016 (C4 (Web Text))
Max Activating Token Index: 369
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 1.4581. Min Range: -1.4581
Max Act: 1.2578. Min Act: -0.0001
Data Index: 1172433 (C4 (Web Text))
Max Activating Token Index: 590
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 1.4581. Min Range: -1.4581
Max Act: 1.3210. Min Act: -0.0001
Data Index: 493760 (C4 (Web Text))
Max Activating Token Index: 237
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 1.4581. Min Range: -1.4581
Max Act: 1.1962. Min Act: -0.0001
Data Index: 817 (C4 (Web Text))
Max Activating Token Index: 619
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 1.4581. Min Range: -1.4581
Max Act: 1.1559. Min Act: -0.0001
Data Index: 219723 (C4 (Web Text))
Max Activating Token Index: 797
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 1.4581. Min Range: -1.4581
Max Act: 1.2946. Min Act: -0.0001
Data Index: 1388020 (Python Code)
Max Activating Token Index: 825
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 1.4581. Min Range: -1.4581
Max Act: 1.1796. Min Act: -0.0001
Data Index: 30891 (C4 (Web Text))
Max Activating Token Index: 771
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 1.4581. Min Range: -1.4581
Max Act: 1.2419. Min Act: -0.0001
Data Index: 1387309 (Python Code)
Max Activating Token Index: 426
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 1.4581. Min Range: -1.4581
Max Act: 1.1254. Min Act: -0.0001
Data Index: 458317 (C4 (Web Text))
Max Activating Token Index: 263
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 1.4581. Min Range: -1.4581
Max Act: 1.1460. Min Act: -0.0001
Data Index: 308132 (C4 (Web Text))
Max Activating Token Index: 580
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 1.4581. Min Range: -1.4581
Max Act: 1.0981. Min Act: -0.0001
Data Index: 1553677 (Python Code)
Max Activating Token Index: 474
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 1.4581. Min Range: -1.4581
Max Act: 1.2493. Min Act: -0.0001
Data Index: 1372749 (Python Code)
Max Activating Token Index: 893
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 1.4581. Min Range: -1.4581
Max Act: 1.1830. Min Act: -0.0001
Data Index: 1440335 (Python Code)
Max Activating Token Index: 601
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 1.4581. Min Range: -1.4581
Max Act: 1.0996. Min Act: -0.0001
Data Index: 1152867 (C4 (Web Text))
Max Activating Token Index: 312
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 1.4581. Min Range: -1.4581
Max Act: 1.1979. Min Act: -0.0001
Data Index: 1027489 (C4 (Web Text))
Max Activating Token Index: 268
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 1.4581. Min Range: -1.4581
Max Act: 1.1625. Min Act: -0.0001
Data Index: 1530386 (Python Code)
Max Activating Token Index: 908
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 1.4581. Min Range: -1.4581
Max Act: 1.2149. Min Act: -0.0001
Data Index: 779738 (C4 (Web Text))
Max Activating Token Index: 258
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 1.4581. Min Range: -1.4581
Max Act: 1.2688. Min Act: -0.0001
Data Index: 436448 (C4 (Web Text))
Max Activating Token Index: 587
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 1.4581. Min Range: -1.4581
Max Act: 1.1105. Min Act: -0.0001
Data Index: 1469058 (Python Code)
Max Activating Token Index: 652
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 1.4581. Min Range: -1.4581
Max Act: 1.1380. Min Act: -0.0001
Data Index: 602407 (C4 (Web Text))
Max Activating Token Index: 597
Click toggle to see full text
Truncated
Full Text #19