© 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. 18057
    Prev
    Next
    INDEX
    Explanations

    graphic / abstract

    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
    natur
    -0.07
    [âĢ¦
    -0.07
    dung
    -0.06
    aum
    -0.06
    fam
    -0.06
    oney
    -0.06
     [~,
    -0.06
    igli
    -0.06
    dur
    -0.06
    ëĭ¤
    -0.06
    POSITIVE LOGITS
    åı£çļĦ
    0.05
    iciens
    0.05
    uge
    0.05
    Ñİ
    0.05
    ary
    0.05
    å¾ij
    0.05
    ÙĦÙĩ
    0.05
     minority
    0.05
    èįł
    0.05
    uvre
    0.05
    Activations Density 0.009%

    No Known Activations