© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 21701
    Prev
    Next
    INDEX
    Explanations

    negatively or decreasing states

    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
    oi
    -0.06
     amen
    -0.06
    okok
    -0.06
     demonstr
    -0.05
     ÙĦÙĪØŃ
    -0.05
    请注æĦı
    -0.05
    usten
    -0.05
    ellingen
    -0.05
    emon
    -0.05
     deliberately
    -0.05
    POSITIVE LOGITS
    akar
    0.06
    issé
    0.05
    èµĦ
    0.05
    acent
    0.05
    åıĬæŶåIJij
    0.05
     برد
    0.05
     Daarna
    0.05
    ån
    0.05
    ä¸įåı¯æĢķ
    0.05
    äºİæĺ¯
    0.05
    Activations Density 0.071%

    No Known Activations