© 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. 34-GEMMASCOPE-RES-131K
    7. 64702
    Prev
    Next
    INDEX
    Explanations

    asking what it means

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

    The neuron fires on the function‐word “skeleton” of explicit questions—especially the “what … does it mean”–style interrogative phrases.

    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
     meaning
    -1.73
    Meaning
    -1.65
    meaning
    -1.64
     Meaning
    -1.63
     which
    -1.05
     signification
    -1.04
     what
    -0.99
     significado
    -0.96
     significato
    -0.93
    作用
    -0.91
    POSITIVE LOGITS
     mean
    1.30
     means
    1.27
     simply
    0.95
    嗎
    0.88
    在你
    0.88
    NewUrlParser
    0.86
     wann
    0.86
    えたら
    0.84
    primarily
    0.84
    Métodos
    0.81
    Activations Density 0.060%

    No Known Activations