© 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. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 22335
    Prev
    Next
    INDEX
    Explanations

    assistance or help requested

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    decoderesearch/qwen-3.5-saes/qwen-3.5-4b
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     may
    -0.06
    ç£Ĭ
    -0.05
    åı¯èĥ½å°±
    -0.05
    íĻĶìĹIJ
    -0.05
    valu
    -0.05
    头æĿ¡
    -0.05
    åĪ°è¿ĩ
    -0.05
     /
    -0.05
    iga
    -0.05
    a
    -0.05
    POSITIVE LOGITS
    ytics
    0.07
    atics
    0.06
    pton
    0.06
    minos
    0.06
     ++$
    0.06
    arze
    0.06
    essos
    0.06
    arau
    0.06
    omod
    0.06
    áticos
    0.06
    Activations Density 0.001%

    No Known Activations