© 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. 25785
    Prev
    Next
    INDEX
    Explanations

    bias voltage, parameters

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

    The neuron fires on relatively uncommon or “rare” content tokens—i.e. low‐frequency words and acronyms.

    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
     as
    -0.88
     inject
    -0.86
    ऴ
    -0.82
     versus
    -0.81
     Vgl
    -0.79
    HasColumnName
    -0.77
    versus
    -0.76
     a
    -0.76
     Patient
    -0.75
     Dat
    -0.75
    POSITIVE LOGITS
    ⠹
    0.98
     tatouage
    0.96
    Jika
    0.95
    adering
    0.94
     flyg
    0.93
     Verpflichtung
    0.92
     arabic
    0.90
     لك
    0.90
    Joe
    0.88
    Craig
    0.88
    Activations Density 0.067%

    No Known Activations