© 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. 22-GEMMASCOPE-RES-131K
    7. 15409
    Prev
    Next
    INDEX
    Explanations

    technical and abstract concepts

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

    The neuron fires on quantitative or measured information—numbers, values, metrics, or statistical data in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    ();?>
    -0.82
     Lungen
    -0.79
     Once
    -0.79
     veľa
    -0.77
     golpea
    -0.77
    estie
    -0.77
    arão
    -0.76
    While
    -0.76
     капю
    -0.76
    During
    -0.75
    POSITIVE LOGITS
     longterm
    0.97
     lignes
    0.95
    imachinery
    0.82
     hao
    0.82
     ild
    0.81
     ROYAL
    0.80
     oke
    0.79
    brothers
    0.79
     Kini
    0.78
     lili
    0.77
    Activations Density 4.184%

    No Known Activations