© 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. 3856
    Prev
    Next
    INDEX
    Explanations

    structured data

    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
     ((__
    -0.23
    ]={↵
    -0.21
    -prev
    -0.20
    presión
    -0.19
    donnees
    -0.19
     Hak
    -0.18
    zure
    -0.18
    LETED
    -0.18
     èĩªåĬ¨çĶŁæĪIJ
    -0.18
    bject
    -0.18
    POSITIVE LOGITS
    ä½£
    0.22
    èµ¢
    0.20
     Decompiled
    0.20
     updateTime
    0.19
    !important
    0.19
    loquent
    0.19
    piler
    0.19
    аблиÑĨ
    0.19
     ground
    0.18
    noc
    0.18
    Activations Density 0.157%

    No Known Activations