© 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. 18266
    Prev
    Next
    INDEX
    Explanations

    see and tell

    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
    rous
    -0.06
    éļĨéļĨ
    -0.06
    am
    -0.06
    å¦Ħ
    -0.06
    endet
    -0.06
    اÙħÙĨ
    -0.06
    hard
    -0.05
    arme
    -0.05
    влаж
    -0.05
    æµ·æ¶Ľ
    -0.05
    POSITIVE LOGITS
    .Comparator
    0.06
    ç¿İ
    0.06
    ogne
    0.06
    ÑĩÑij
    0.06
    ç½
    0.06
    itis
    0.06
     Hert
    0.06
    asking
    0.05
    urn
    0.05
    ience
    0.05
    Activations Density 0.004%

    No Known Activations