© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 93507
    Prev
    Next
    INDEX
    Explanations

    conditional phrases starting with if

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

    The neuron specializes in spotting code‐structure markers—things like colons, braces/brackets, indentation tokens, and similar punctuation that indicate source‐code formatting.

    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
     alltså
    -1.88
     verkligen
    -1.88
     actualizaciones
    -1.79
     dessutom
    -1.76
     samtidigt
    -1.76
    いいのか
    -1.74
     säga
    -1.74
     faktiskt
    -1.73
     ändå
    -1.72
     långt
    -1.71
    POSITIVE LOGITS
     if
    13.75
     если
    8.06
    if
    7.25
     If
    6.84
     якщо
    6.28
     nếu
    6.13
     اگر
    6.09
    If
    5.88
     dacă
    5.78
     jika
    5.53
    Activations Density 1.379%

    No Known Activations