© 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. 112559
    Prev
    Next
    INDEX
    Explanations

    one [adjective] [noun]

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

    The neuron activates on degree-expressing qualifiers (intensifiers) such as “huge,” “very,” “big,” “common,” and similar words that amplify or emphasize.

    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
     to
    -1.87
    躂
    -1.75
    裥
    -1.63
    
    -1.62
     themselves
    -1.59
    zule
    -1.49
    就行了
    -1.48
     لأن
    -1.46
     TMPro
    -1.45
     estadounid
    -1.44
    POSITIVE LOGITS
    netics
    1.97
    (!__
    1.82
     onOptions
    1.78
     GenerationType
    1.77
    ühnen
    1.77
     September
    1.73
    adone
    1.69
    figer
    1.68
     forState
    1.67
    naments
    1.66
    Activations Density 0.033%

    No Known Activations