© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 75485
    Prev
    Next
    INDEX
    Explanations

    cultivation, renovation, excavation, cultivar

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

    The neuron detects the subword “culti,” i.e. occurrences of the root “culti” in words (cultivable, cultivated, cultivation, uncultivated, cultivar, 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_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
     replays
    -0.69
    缮
    -0.69
     TM
    -0.68
     relive
    -0.68
    三浦
    -0.68
     rechts
    -0.67
     gestorben
    -0.65
    ┆
    -0.64
    DD
    -0.63
     undergo
    -0.63
    POSITIVE LOGITS
     baseman
    0.90
    文明
    0.82
    <!--[
    0.81
    ateness
    0.79
    iffa
    0.77
    baiki
    0.76
    ated
    0.74
    цията
    0.74
     keter
    0.73
    edale
    0.73
    Activations Density 0.037%

    No Known Activations