© 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. 34-GEMMASCOPE-RES-131K
    7. 57210
    Prev
    Next
    INDEX
    Explanations

    Madonna and Child

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

    The neuron strongly activates on long runs of leading or repeated whitespace (i.e. indentation tokens).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    πάρχ
    -0.83
     unbelie
    -0.80
     paternity
    -0.79
     पा
    -0.77
     hashtag
    -0.76
     मान
    -0.76
     mandato
    -0.75
     Meilleur
    -0.74
     juveniles
    -0.73
     figuras
    -0.73
    POSITIVE LOGITS
     Madonna
    1.15
    Madonna
    1.14
     Malawi
    0.94
     madonna
    0.94
    Malawi
    0.90
    balah
    0.88
     Kab
    0.88
    Kab
    0.86
     Novi
    0.85
     mdi
    0.84
    Activations Density 0.005%

    No Known Activations