© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 1586
    Prev
    Next
    INDEX
    Explanations

    database query results

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

    The neuron preferentially activates on numerical tokens (digits, punctuation, and parentheses), especially in version numbers or other digit‐rich 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
     ren
    -0.91
    ISER
    -0.78
     notar
    -0.77
     conten
    -0.75
     nom
    -0.74
    BUR
    -0.73
     xxx
    -0.73
    alus
    -0.71
    Ae
    -0.71
     blan
    -0.70
    POSITIVE LOGITS
     Kram
    0.91
     Zat
    0.77
     rebuilding
    0.77
     tulis
    0.77
    siyon
    0.76
     esperança
    0.75
    долго
    0.75
    reef
    0.75
     Frontera
    0.73
     ketiga
    0.73
    Activations Density 0.014%

    No Known Activations