© 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. 89751
    Prev
    Next
    INDEX
    Explanations

    code and variations of "can"

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

    The neuron is detecting individual digit tokens (especially “0”) in code‐like or numeric sequences.

    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
    Dalam
    -0.99
    Henry
    -0.96
    𝐑
    -0.94
     Receptors
    -0.92
     DatePicker
    -0.91
    Hình
    -0.91
    laca
    -0.90
     having
    -0.89
    Jeśli
    -0.88
    قامت
    -0.88
    POSITIVE LOGITS
     molte
    0.94
     verschiedene
    0.94
     diversos
    0.93
     verschillende
    0.90
     verschiedenen
    0.88
     alguns
    0.85
    Funded
    0.85
     alcune
    0.84
    Versatile
    0.81
    xtra
    0.81
    Activations Density 0.004%

    No Known Activations