© 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. 10-GEMMASCOPE-RES-131K
    7. 45191
    Prev
    Next
    INDEX
    Explanations

    introducing ways or means

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

    This neuron specifically lights up on the token “Through,” detecting occurrences of that word (often at the start of phrases).

    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
     sonriendo
    -1.84
     when
    -1.84
     other
    -1.67
     schnellen
    -1.66
     زیرا
    -1.65
     الأخرى
    -1.63
     sosteniendo
    -1.63
     as
    -1.61
     is
    -1.59
     use
    -1.59
    POSITIVE LOGITS
    ⑊
    1.89
    瞜
    1.79
    ࿚
    1.77
     Conſ
    1.72
    Ihre
    1.69
     הם
    1.67
     kaas
    1.66
     Inſ
    1.64
    Through
    1.62
    ꜆
    1.59
    Activations Density 0.009%

    No Known Activations