© 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. 16954
    Prev
    Next
    INDEX
    Explanations

    multiple factors

    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
    å¤ļåıĺ
    -0.06
    åıĤ
    -0.05
    um
    -0.05
    le
    -0.05
    å¦Ĥå®ŀ
    -0.05
    amb
    -0.05
    ob
    -0.05
    loch
    -0.05
     reinterpret
    -0.05
    æ··
    -0.05
    POSITIVE LOGITS
    رÙĬج
    0.06
    attr
    0.06
    junto
    0.06
    förderung
    0.06
     attr
    0.06
    ilÉĻr
    0.06
    ä¸ŃæĭĽ
    0.06
    缺ä¸Ģä¸įåı¯
    0.06
    éķ¿æķĪæľºåĪ¶
    0.05
    ODA
    0.05
    Activations Density 0.016%

    No Known Activations