© 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. 34-GEMMASCOPE-RES-131K
    7. 2885
    Prev
    Next
    INDEX
    Explanations

    names ending in ovich or ova

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

    The neuron is attuned to specific short letter‐sequences (character trigrams)—like “ll,” “old,” “ral,” “ing,” “se,” “lu,” “ach,” etc.—that frequently appear as parts of personal and place names.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    יט
    -0.71
     wyra
    -0.70
    𝐀
    -0.68
    genheit
    -0.68
    Rider
    -0.68
     vengeance
    -0.68
    にしても
    -0.67
     fallait
    -0.67
    𝖚
    -0.66
    Flan
    -0.66
    POSITIVE LOGITS
     settings
    0.73
     рисо
    0.72
     Judith
    0.71
    unlike
    0.70
    closest
    0.69
    VISA
    0.69
     Aarhus
    0.69
    릎
    0.68
    Settings
    0.68
     واب
    0.68
    Activations Density 0.057%

    No Known Activations