© 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. 112398
    Prev
    Next
    INDEX
    Explanations

    prepositions followed by nouns/adjectives

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

    The neuron activates on common prepositions and prepositional cues (like “under,” “in,” “by,” “using,” “on,” “across,” “upon,” etc.) that introduce experimental conditions or methodological context.

    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
    튿
    -1.52
    Bibliografie
    -1.40
     takže
    -1.40
    ühle
    -1.39
     quantità
    -1.39
    を書く
    -1.38
    の詳細
    -1.38
    这篇
    -1.35
    も含
    -1.34
    PARTE
    -1.34
    POSITIVE LOGITS
     which
    1.96
     an
    1.68
     from
    1.65
     whose
    1.65
     that
    1.56
     his
    1.54
     whom
    1.54
     Their
    1.53
     their
    1.51
     who
    1.49
    Activations Density 0.376%

    No Known Activations