© 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. 90367
    Prev
    Next
    INDEX
    Explanations

    specific outcomes and qualities

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

    The neuron fires strongly on long, multi‐syllable Latin‐derived roots (e.g. “scientif-,” “magnif-,” “signif-,” “sacrile-,” “por-” as in “specific,” “malef-” as in “maleficent,” 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
     inelastic
    -0.69
     mese
    -0.66
    予想
    -0.65
    OTHY
    -0.65
    óry
    -0.65
     occurring
    -0.65
     ultimatum
    -0.63
     byproduct
    -0.63
    らしく
    -0.63
     Tianjin
    -0.63
    POSITIVE LOGITS
    ific
    1.66
    IFIC
    1.30
    fic
    1.16
     specificity
    0.99
     especific
    0.97
    specific
    0.94
    pecific
    0.91
    SPECIFIC
    0.90
     spécifique
    0.86
    efic
    0.85
    Activations Density 0.039%

    No Known Activations