© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 19928
    Prev
    Next
    INDEX
    Explanations

    hottest followed by topics

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

    The neuron fires on the adjective “hot” (especially its superlative form “hottest”) when it’s used to hype or trend-label something.

    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
    The
    -1.99
    ohn
    -1.88
    u
    -1.82
    م
    -1.78
    (
    -1.71
    =’
    -1.66
    εια
    -1.64
    お前
    -1.63
    Its
    -1.62
    In
    -1.60
    POSITIVE LOGITS
    硨
    1.80
     🔥
    1.76
     cuentan
    1.61
     causando
    1.59
     такая
    1.57
     tratando
    1.55
     ―
    1.53
     [
    1.51
    棡
    1.51
     comienzan
    1.49
    Activations Density 0.019%

    No Known Activations