© 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. 19878
    Prev
    Next
    INDEX
    Explanations

    struct or class 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
    为ä¹ĭ
    -0.06
    udit
    -0.06
    rapy
    -0.05
    mo
    -0.05
    诡
    -0.05
    被åıijçİ°
    -0.05
    纳
    -0.05
    мени
    -0.05
    tod
    -0.05
     Bong
    -0.05
    POSITIVE LOGITS
    çĶŁåľ¨
    0.06
    ovar
    0.06
    éľĢè¦ģ对
    0.06
    vara
    0.05
    éľĢè¦ģåľ¨
    0.05
    Ùĩاد
    0.05
    à¸ģลาà¸ĩ
    0.05
    egin
    0.05
    rary
    0.05
     sinc
    0.05
    Activations Density 0.004%

    No Known Activations