© 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. Qwen2.5-7B-IT
    5. 20-MATRYOSHKA-65K
    7. 1547
    Prev
    Next
    INDEX
    Explanations

    how, why, impact

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    chanind/qwen2.5-7B-it-layer-20-saes/pile/matryoshka/k-100
    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
     legalize
    -0.26
     resembl
    -0.23
    å°±æĿ¥çľĭçľĭ
    -0.23
    ä¸Ģèĩ´å¥½è¯Ħ
    -0.23
     tempList
    -0.23
    ãİı
    -0.23
    âĢ½
    -0.22
    åīįæıIJæĺ¯
    -0.22
    éĤ£ä¹Ī容æĺĵ
    -0.21
     à¸ĺ
    -0.21
    POSITIVE LOGITS
     how
    0.46
    å¦Ĥä½ķ
    0.43
    how
    0.37
    æĺ¯å¦Ĥä½ķ
    0.36
     possible
    0.33
     behavior
    0.31
     impact
    0.31
     relationship
    0.29
     why
    0.28
    æĢİæł·
    0.28
    Activations Density 6.092%

    No Known Activations