© 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. 66904
    Prev
    Next
    INDEX
    Explanations

    prefixing words

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

    The neuron is picking out numeric literals embedded in source‐code snippets (e.g. the “80” version constant in the C# method).

    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
    sprüng
    -1.19
    merz
    -1.19
     scoperto
    -1.18
    ּוֹ
    -1.18
    ācija
    -1.14
     minuter
    -1.12
     pantalon
    -1.09
    🫑
    -1.07
    vább
    -1.07
     Informações
    -1.06
    POSITIVE LOGITS
     сахара
    1.13
     and
    1.07
    も行
    1.02
     ibland
    1.00
    onaldo
    0.97
     ویتامین
    0.97
     présence
    0.96
     zudem
    0.96
     samtidigt
    0.96
    umque
    0.96
    Activations Density 0.000%

    No Known Activations