© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 117811
    Prev
    Next
    INDEX
    Explanations

    points and domain

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

    The neuron fires on the opening question phrase of a math problem—i.e. on sentence-initial interrogative words like “What,” “Are,” or “Limits” that introduce a 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
     напо
    -0.74
     during
    -0.74
     fondo
    -0.71
     Segu
    -0.70
     lanza
    -0.67
    orros
    -0.65
     เก
    -0.65
    Ẫ
    -0.65
    toires
    -0.65
    cropped
    -0.65
    POSITIVE LOGITS
     ordered
    1.19
     point
    1.11
     coordinate
    1.04
     points
    1.02
    ordered
    1.01
     Ordered
    0.94
     Quadrant
    0.94
    坐标
    0.94
     Slope
    0.93
     Coordinates
    0.92
    Activations Density 0.026%

    No Known Activations