© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3.5-9B Base
    5. 15-QWENSCOPE-RES-64K
    7. 672
    Prev
    Next
    INDEX
    Explanations

    separation and selective removal

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
    angat
    -0.07
    åĩĮ
    -0.07
    ÙĨب
    -0.07
    .va
    -0.07
    iosi
    -0.07
    ä¸ĢåŃĹ
    -0.06
     Inst
    -0.06
     PROP
    -0.06
    è·¨è¶Ĭ
    -0.06
    æľ«å°¾
    -0.06
    POSITIVE LOGITS
     separating
    0.12
    åĪĨ离
    0.12
     separation
    0.12
     selectively
    0.11
     Separ
    0.10
    Selective
    0.10
     selective
    0.09
    çļĦå·®å¼Ĥ
    0.09
     separates
    0.09
     discriminate
    0.09
    Activations Density 0.045%

    No Known Activations