© 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. 10-GEMMASCOPE-RES-131K
    7. 84602
    Prev
    Next
    INDEX
    Explanations

    incremental changes or additions

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

    The neuron fires strongly on the word “incremental” (and its subword pieces).

    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
    “
    -3.31
     uly
    -3.03
    ”.
    -2.94
    鈈
    -2.81
    ኧ
    -2.77
    =“
    -2.66
    {
    -2.50
     hii
    -2.47
    ”،
    -2.45
     kush
    -2.41
    POSITIVE LOGITS
     fueron
    2.97
     «
    2.64
     :
    2.56
     distraught
    2.39
    轶
    2.38
    取締
    2.31
    文中
    2.30
    сибир
    2.28
     ziehen
    2.22
    郸
    2.22
    Activations Density 0.002%

    No Known Activations