© 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. Qwen2.5-7B-IT
    5. 20-MATRYOSHKA-65K
    7. 3604
    Prev
    Next
    INDEX
    Explanations

    code path configuration

    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
     Marketable
    -0.19
    repid
    -0.19
    åĸ½
    -0.18
    æ¹İ
    -0.18
    se
    -0.18
    è¿«ä¸įåıĬå¾ħ
    -0.18
     bindActionCreators
    -0.18
     adulti
    -0.18
    å®īåħ¨ä¿Ŀéļľ
    -0.18
    noxious
    -0.18
    POSITIVE LOGITS
    -scalable
    0.26
    errorMsg
    0.24
    thinkable
    0.24
    même
    0.23
    /Instruction
    0.22
     negó
    0.21
    DMETHOD
    0.21
    á¿ĩ
    0.21
     newcom
    0.21
    forgettable
    0.21
    Activations Density 0.906%

    No Known Activations