© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 129374
    Prev
    Next
    INDEX
    Explanations

    preposition followed by place or institution

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

    This neuron is selectively activating on numeric values (e.g. years, counts, measurements) appearing in the 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_10/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
    Karakteristik
    -2.05
     realmente
    -1.76
    獠
    -1.57
     последних
    -1.50
    が無い
    -1.47
    が少ない
    -1.45
     môžete
    -1.45
     musíte
    -1.42
     normalmente
    -1.41
     těch
    -1.39
    POSITIVE LOGITS
     and
    2.27
     by
    2.27
     a
    2.06
    .”
    1.62
     two
    1.59
     three
    1.59
     what
    1.55
     to
    1.55
     but
    1.48
     of
    1.45
    Activations Density 0.339%

    No Known Activations