© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 67078
    Prev
    Next
    INDEX
    Explanations

    id, name, comparison

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

    This neuron activates on numeric literals (stand-alone numbers) 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
    fang
    -0.88
     Associated
    -0.88
    why
    -0.83
    grading
    -0.82
    诡异
    -0.79
    挫
    -0.79
    achable
    -0.79
    ʹ
    -0.78
     oltre
    -0.77
     '-';
    -0.76
    POSITIVE LOGITS
     comparison
    1.08
     comparer
    0.95
     comparing
    0.85
    なんですよ
    0.85
     comparator
    0.84
    ansı
    0.82
     compareTo
    0.82
     jsi
    0.81
     funkcji
    0.81
     Kč
    0.81
    Activations Density 0.005%

    No Known Activations