© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 70250
    Prev
    Next
    INDEX
    Explanations

    Hil names and places

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

    The neuron activates on occurrences of the substring “Hil” (e.g. in names or terms like Hilbert, Hildegard, Hilversum, etc.).

    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
     proves
    -0.96
     prove
    -0.80
     proved
    -0.78
    proven
    -0.76
     substantiate
    -0.75
     adicionais
    -0.75
     元素
    -0.75
     Prove
    -0.75
     proving
    -0.74
    來自
    -0.73
    POSITIVE LOGITS
     HIL
    1.13
     Hil
    0.96
    HIL
    0.94
    quias
    0.91
     Hild
    0.85
     Nuk
    0.85
    brand
    0.81
     HUND
    0.80
     HOL
    0.78
     ヒ
    0.78
    Activations Density 0.011%

    No Known Activations