© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 29070
    Prev
    Next
    INDEX
    Explanations

    Me or It

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

    The neuron fires on short, two-letter tokens like “me,” “it,” or “up” (often in UI labels or tiny command/function names).

    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
     each
    -1.91
    Ꮬ
    -1.66
    Ꮉ
    -1.62
    Beschrijving
    -1.55
    fitriones
    -1.55
     każdy
    -1.52
     ideen
    -1.52
     slutet
    -1.49
    ⓪
    -1.48
    𨧀
    -1.48
    POSITIVE LOGITS
     It
    1.84
    莛
    1.67
    This
    1.52
    茀
    1.52
     Not
    1.45
    ennessee
    1.42
     вторая
    1.40
     posiblemente
    1.38
    苼
    1.38
    
    1.38
    Activations Density 0.037%

    No Known Activations