© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 21684
    Prev
    Next
    INDEX
    Explanations

    software build paths

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

    The neuron is keyed to numeric tokens—especially numbers appearing in file paths or build‐directory references.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    ien
    -1.89
    BindingResult
    -1.83
    tellung
    -1.80
    argu
    -1.80
    -1.77
    Ꮴ
    -1.77
    馐
    -1.77
    ⵌ
    -1.76
    hip
    -1.75
    e
    -1.73
    POSITIVE LOGITS
    build
    2.34
     If
    2.28
    )?
    2.13
    .”.
    2.06
    .”)
    2.03
    Odpovědět
    2.02
     Build
    1.94
     We
    1.94
    也
    1.92
     construcción
    1.90
    Activations Density 0.029%

    No Known Activations