© 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. 33184
    Prev
    Next
    INDEX
    Explanations

    default values and settings

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

    The neuron fires on numeric literals and related tokens (integers, floats, constants) in code.

    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
    ญี่ปุ่น
    -0.89
     bête
    -0.85
    IAm
    -0.84
     marcó
    -0.84
     gila
    -0.82
     pistolet
    -0.82
    формы
    -0.81
     Rè
    -0.81
     Vedi
    -0.80
     vigilante
    -0.79
    POSITIVE LOGITS
     default
    1.92
    getDefault
    1.74
    default
    1.71
    DEFAULT
    1.69
    Default
    1.69
     automatically
    1.68
     Default
    1.66
     getDefault
    1.58
     DEFAULT
    1.52
     automatic
    1.48
    Activations Density 0.028%

    No Known Activations