© 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. 318
    Prev
    Next
    INDEX
    Explanations

    string concatenation with +

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

    This neuron activates on numeric literals (numbers) in code or text.

    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
    kyt
    -0.82
    啼
    -0.81
    сыз
    -0.80
     bénéficie
    -0.78
    濡
    -0.76
     utilizó
    -0.75
    стан
    -0.74
    不大
    -0.74
    Variance
    -0.73
     those
    -0.72
    POSITIVE LOGITS
     str
    1.66
    str
    1.49
     stratified
    1.19
    stris
    0.94
    ⫹
    0.92
    lifies
    0.90
    imentos
    0.85
     aste
    0.85
     orific
    0.85
     illustrationer
    0.85
    Activations Density 0.004%

    No Known Activations