© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 3341
    Prev
    Next
    INDEX
    Explanations

    how to do

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

    The neuron fires on the interrogative “How” (and its accompanying question‐intro words like “do,” “I,” “can”) that signal the start of a user’s question.

    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
     have
    -1.00
     것이다
    -0.90
    getMinutes
    -0.89
     trampa
    -0.88
     haben
    -0.88
    ungszeit
    -0.84
    tieve
    -0.83
     tránsito
    -0.83
    SetOutput
    -0.82
    rophore
    -0.82
    POSITIVE LOGITS
     can
    2.17
     to
    1.82
     do
    1.71
     можно
    1.16
     podemos
    1.16
     galite
    1.00
     best
    0.98
     môže
    0.97
     unaweza
    0.96
     possono
    0.93
    Activations Density 0.084%

    No Known Activations