© 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. 21154
    Prev
    Next
    INDEX
    Explanations

    MS or ms

    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
    erta
    -0.06
     contr
    -0.05
    é
    -0.05
     sost
    -0.05
    oe
    -0.05
     importantly
    -0.05
    sw
    -0.05
    l
    -0.04
    åĮºåĪĨ
    -0.04
    J
    -0.04
    POSITIVE LOGITS
    oarece
    0.07
    icante
    0.07
    naver
    0.07
    мееÑĤÑģÑı
    0.07
    evin
    0.07
    æŶåĢĻ
    0.07
    peration
    0.07
    PTS
    0.06
    atorów
    0.06
     confia
    0.06
    Activations Density 0.004%

    No Known Activations