© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 20562
    Prev
    Next
    INDEX
    Explanations

    prefix for common suffixes

    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
    s
    -0.07
    c
    -0.06
    Äĥ
    -0.05
     ing
    -0.05
    anger
    -0.05
    ERA
    -0.05
    ariana
    -0.05
     Z
    -0.05
    ãģĨ
    -0.05
    v
    -0.05
    POSITIVE LOGITS
    epend
    0.06
    ezas
    0.06
    дин
    0.06
    chaffen
    0.06
    qués
    0.06
    大ä¸ļ
    0.06
    дим
    0.06
    erras
    0.06
    ainty
    0.06
    fras
    0.06
    Activations Density 0.601%

    No Known Activations