© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 59751
    Prev
    Next
    INDEX
    Explanations

    onon

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

    The neuron strongly activates on gerund or –ing forms (and related nouns) that appear in temporal prepositional phrases—especially constructions like “on returning,” “on coming,” “on clicking,” “upon completing,” etc.

    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
    婶
    -1.55
     same
    -1.53
     until
    -1.46
     options
    -1.45
    ارنة
    -1.41
     excepcional
    -1.38
     occasions
    -1.38
     है
    -1.35
     antes
    -1.34
    )});
    -1.34
    POSITIVE LOGITS
     groet
    1.40
     Misalnya
    1.39
     hôtes
    1.31
    𝗪
    1.29
     fortu
    1.29
    穣
    1.29
     tyrants
    1.28
     fidèles
    1.28
     عندما
    1.27
     begre
    1.27
    Activations Density 0.044%

    No Known Activations