© 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. 85442
    Prev
    Next
    INDEX
    Explanations

    accompanying or in sequence

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

    This neuron responds to idiomatic multiword expressions of accompaniment (e.g. “hand in hand,” “go along with”).

    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
    ܊
    -1.73
     nových
    -1.70
    Ꮛ
    -1.70
     is
    -1.66
    piej
    -1.64
     هناك
    -1.59
     zakup
    -1.57
     \_
    -1.57
     nowych
    -1.57
    competent
    -1.56
    POSITIVE LOGITS
     then
    2.13
    みたいに
    1.61
     obliga
    1.55
    O
    1.53
    も行
    1.53
     italiani
    1.51
     aims
    1.48
     like
    1.47
    and
    1.45
    になりました
    1.45
    Activations Density 0.003%

    No Known Activations