© 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. 113548
    Prev
    Next
    INDEX
    Explanations

    locations and organizations

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

    The neuron is keyed to very common short function words (e.g. frequent prepositions and conjunctions such as “in,” “and,” or citation fragments like “et al”).

    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
     -
    -1.78
    vamo
    -1.73
    On
    -1.66
     also
    -1.63
     prickly
    -1.62
    2
    -1.60
    没想到
    -1.58
     will
    -1.58
     將
    -1.58
     was
    -1.56
    POSITIVE LOGITS
     the
    2.23
    鬻
    2.02
    _
    1.98
    幛
    1.98
     vaak
    1.98
     münchen
    1.92
    糴
    1.89
     konkrét
    1.82
    粜
    1.80
     devenus
    1.80
    Activations Density 0.031%

    No Known Activations