© 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. 22-GEMMASCOPE-RES-131K
    7. 39914
    Prev
    Next
    INDEX
    Explanations

    code snippets and structure

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

    This neuron fires on code identifiers for data‐type or class names (e.g. CamelCase/PascalCase type tokens like DirectoryInfo, FileInfo, PlaceOrderDto, Length).

    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
    孕妇
    -0.91
    atuur
    -0.88
    詆
    -0.84
     majú
    -0.83
    當初
    -0.81
    esome
    -0.81
    につなが
    -0.81
    berichte
    -0.80
    wände
    -0.80
     suscripción
    -0.79
    POSITIVE LOGITS
     existing
    1.57
     already
    1.54
     уже
    1.12
     bereits
    1.11
    OrCreate
    1.05
    existing
    1.05
     EXISTING
    1.04
     almost
    1.02
     esist
    1.02
     new
    1.01
    Activations Density 0.032%

    No Known Activations