© 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. 18152
    Prev
    Next
    INDEX
    Explanations

    important to perform

    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
     accordingly
    -0.07
    uck
    -0.06
    ç¦ıå¾·
    -0.05
    ly
    -0.05
    like
    -0.05
    ä»»ä½ķ
    -0.05
     it
    -0.05
     if
    -0.05
    gon
    -0.05
    nd
    -0.05
    POSITIVE LOGITS
     terce
    0.07
    ä¸įæĬĬ
    0.07
    andr
    0.07
    quio
    0.06
    以ä¸Ĭæĺ¯
    0.06
    gens
    0.06
    rensa
    0.06
    jaw
    0.06
    aname
    0.06
    æĬĬè¿Ļ个
    0.06
    Activations Density 0.006%

    No Known Activations