© 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. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 15717
    Prev
    Next
    INDEX
    Explanations

    code comments and definitions

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron responds to numeric literals—especially floating‐point or version numbers with decimal points.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Zobacz
    -1.01
     Bucure
    -0.94
    彼は
    -0.93
     campesinos
    -0.89
    ён
    -0.85
     București
    -0.85
     έχουν
    -0.85
    さらに
    -0.84
    Salut
    -0.84
    かわらず
    -0.84
    POSITIVE LOGITS
    mml
    0.98
    bbage
    0.97
    utorial
    0.94
    人也
    0.91
     recomendado
    0.91
    等人
    0.89
    ยน
    0.89
    塀
    0.89
    cashire
    0.88
    嚕
    0.88
    Activations Density 0.001%

    No Known Activations