© 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. 85249
    Prev
    Next
    INDEX
    Explanations

    component of

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

    The neuron fires on phrases that state something is a key/important part or component of something (e.g. “important part of,” “essential component of”).

    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
     físicos
    -1.76
     teď
    -1.74
     similaire
    -1.73
    具体的な
    -1.66
     někdo
    -1.63
    が少ない
    -1.63
     señala
    -1.55
    řízen
    -1.55
    ꒦
    -1.52
    formatics
    -1.51
    POSITIVE LOGITS
    re
    2.25
    le
    2.17
     and
    1.91
     their
    1.77
     will
    1.71
    T
    1.71
    man
    1.66
    j
    1.66
     our
    1.66
    ka
    1.64
    Activations Density 0.039%

    No Known Activations