© 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. 34-GEMMASCOPE-RES-131K
    7. 31668
    Prev
    Next
    INDEX
    Explanations

    hill and mountain features

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

    The neuron consistently lights up on numeric tokens—years, measurements, heights, and other digit sequences—in running text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    ierać
    -0.82
    picasso
    -0.79
    LICENSE
    -0.79
    orghe
    -0.79
    tispiece
    -0.77
    μεν
    -0.75
    閤
    -0.75
    ultra
    -0.73
    BOTTOM
    -0.73
     permiso
    -0.73
    POSITIVE LOGITS
     hill
    3.02
     hills
    2.52
     Hill
    2.05
    hill
    1.94
     HILL
    1.88
    Hill
    1.82
    hills
    1.72
     хол
    1.65
    HILL
    1.61
    Hills
    1.59
    Activations Density 0.042%

    No Known Activations