© 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. 20978
    Prev
    Next
    INDEX
    Explanations

    major findings and events

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

    This neuron responds to occurrences of the phrase “major new,” as in headlines or titles announcing a significant new development.

    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
    ,"
    -1.52
    “
    -1.52
     dañ
    -1.52
    ',
    -1.51
    возмо
    -1.50
     prescriptions
    -1.48
     béisbol
    -1.48
     amenaz
    -1.46
    温暖
    -1.45
     ofrec
    -1.45
    POSITIVE LOGITS
    これだけ
    1.59
     oxu
    1.57
    major
    1.56
    <bos>
    1.53
    遊ぶ
    1.51
    あっという
    1.50
    落ち着いた
    1.49
     boya
    1.48
     castig
    1.48
    ◆◆
    1.47
    Activations Density 0.068%

    No Known Activations