© 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. 20403
    Prev
    Next
    INDEX
    Explanations

    means, ratio, form, structure, patterns

    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
     modalitÃł
    -0.06
     gig
    -0.06
     authent
    -0.06
    kind
    -0.06
    ó
    -0.06
    éĤ£ç§į
    -0.06
    模å¼ıåĴĮ
    -0.06
    æ¸ħ
    -0.06
    صÙĬÙĦ
    -0.06
     faithful
    -0.06
    POSITIVE LOGITS
    ussy
    0.06
    asures
    0.06
     ratio
    0.06
     uncon
    0.05
    å¤Ħæĸ¹
    0.05
    è¡¡éĩı
    0.05
     an
    0.05
     weighted
    0.05
    ratio
    0.05
     MIG
    0.05
    Activations Density 0.012%

    No Known Activations