© 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. 17778
    Prev
    Next
    INDEX
    Explanations

    Watson names

    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
     curs
    -0.06
    aniel
    -0.06
    ÙĦÙī
    -0.05
    梢
    -0.05
    /blob
    -0.05
    arn
    -0.05
    uchs
    -0.05
     end
    -0.05
    anel
    -0.05
    磫
    -0.05
    POSITIVE LOGITS
    第ä¸Ģåį°è±¡
    0.06
     Daya
    0.06
    acken
    0.06
     adéqu
    0.06
    mien
    0.06
    ØŃÙħاÙħ
    0.06
    ivor
    0.06
    -quote
    0.06
     Applicant
    0.06
    ÙijÙĩ
    0.06
    Activations Density 0.001%

    No Known Activations