Model: SoLU Model: 10 Layers, 5120 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 4321 in Layer 4
Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')
Text #0
Max Range: 1.8143. Min Range: -1.8143
Max Act: 1.8143. Min Act: -0.0001
Data Index: 630028 (C4 (Web Text))
Max Activating Token Index: 403
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 1.8143. Min Range: -1.8143
Max Act: 1.6489. Min Act: -0.0001
Data Index: 291025 (C4 (Web Text))
Max Activating Token Index: 721
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 1.8143. Min Range: -1.8143
Max Act: 1.7307. Min Act: -0.0001
Data Index: 328731 (C4 (Web Text))
Max Activating Token Index: 165
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 1.8143. Min Range: -1.8143
Max Act: 1.6770. Min Act: -0.0001
Data Index: 675726 (C4 (Web Text))
Max Activating Token Index: 403
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 1.8143. Min Range: -1.8143
Max Act: 1.6680. Min Act: -0.0001
Data Index: 477442 (C4 (Web Text))
Max Activating Token Index: 938
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 1.8143. Min Range: -1.8143
Max Act: 1.6571. Min Act: -0.0001
Data Index: 1467773 (Python Code)
Max Activating Token Index: 321
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 1.8143. Min Range: -1.8143
Max Act: 1.5341. Min Act: -0.0001
Data Index: 1340928 (C4 (Web Text))
Max Activating Token Index: 28
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 1.8143. Min Range: -1.8143
Max Act: 1.5564. Min Act: -0.0001
Data Index: 772557 (C4 (Web Text))
Max Activating Token Index: 815
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 1.8143. Min Range: -1.8143
Max Act: 1.4918. Min Act: -0.0001
Data Index: 982428 (C4 (Web Text))
Max Activating Token Index: 502
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 1.8143. Min Range: -1.8143
Max Act: 1.4466. Min Act: -0.0001
Data Index: 903513 (C4 (Web Text))
Max Activating Token Index: 208
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 1.8143. Min Range: -1.8143
Max Act: 1.5588. Min Act: -0.0001
Data Index: 570390 (C4 (Web Text))
Max Activating Token Index: 812
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 1.8143. Min Range: -1.8143
Max Act: 1.3590. Min Act: -0.0001
Data Index: 177273 (C4 (Web Text))
Max Activating Token Index: 11
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 1.8143. Min Range: -1.8143
Max Act: 1.2779. Min Act: -0.0001
Data Index: 1291739 (C4 (Web Text))
Max Activating Token Index: 43
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 1.8143. Min Range: -1.8143
Max Act: 1.3216. Min Act: -0.0001
Data Index: 1180104 (C4 (Web Text))
Max Activating Token Index: 266
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 1.8143. Min Range: -1.8143
Max Act: 1.2208. Min Act: -0.0001
Data Index: 792179 (C4 (Web Text))
Max Activating Token Index: 360
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 1.8143. Min Range: -1.8143
Max Act: 1.2611. Min Act: -0.0001
Data Index: 754023 (C4 (Web Text))
Max Activating Token Index: 223
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 1.8143. Min Range: -1.8143
Max Act: 1.2114. Min Act: -0.0001
Data Index: 58445 (C4 (Web Text))
Max Activating Token Index: 348
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 1.8143. Min Range: -1.8143
Max Act: 1.1991. Min Act: -0.0001
Data Index: 715500 (C4 (Web Text))
Max Activating Token Index: 713
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 1.8143. Min Range: -1.8143
Max Act: 1.0655. Min Act: -0.0001
Data Index: 1253234 (C4 (Web Text))
Max Activating Token Index: 69
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 1.8143. Min Range: -1.8143
Max Act: 0.9371. Min Act: -0.0001
Data Index: 306459 (C4 (Web Text))
Max Activating Token Index: 961
Click toggle to see full text
Truncated
Full Text #19