© 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. 34-GEMMASCOPE-RES-131K
    4. 82472
    Prev
    Next
    INDEX
    Explanations

    key "id" or "version"

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

    This neuron highly activates on indentation tokens (runs of leading spaces/tabs) in code or structured data.

    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
    などが
    -0.83
     monday
    -0.81
     غير
    -0.81
     Forgot
    -0.78
     Bestand
    -0.78
    laar
    -0.77
    isert
    -0.77
    tellbar
    -0.76
     anpassen
    -0.75
     Monday
    -0.75
    POSITIVE LOGITS
     which
    0.86
    関連リンク
    0.86
    rodz
    0.86
    ORIGINAL
    0.86
     Frucht
    0.83
    grü
    0.82
     Steinberg
    0.81
    Autore
    0.81
    ⟶
    0.80
     objekt
    0.79
    Activations Density 0.099%

    No Known Activations