© 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. 6775
    Prev
    Next
    INDEX
    Explanations

    annual publications and classifications

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

    The neuron detects bibliographic citation markers—especially journal-name tokens starting with “Ann…” (e.g. Ann., Annual, Annu., Annals, Annular, 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
     länder
    -0.84
     головой
    -0.83
     available
    -0.82
     nervosa
    -0.77
     partecip
    -0.75
    🫲
    -0.75
    rén
    -0.74
     製
    -0.73
     muuta
    -0.73
    ölcs
    -0.71
    POSITIVE LOGITS
     of
    0.80
     reflecting
    0.75
    jab
    0.75
     NY
    0.73
     combust
    0.73
     guardians
    0.73
    міст
    0.72
    阳
    0.71
    Reflect
    0.71
    comb
    0.69
    Activations Density 0.023%

    No Known Activations