© 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. 22-GEMMASCOPE-RES-131K
    4. 36552
    Prev
    Next
    INDEX
    Explanations

    this is a phrase starter

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

    The neuron fires on numeric sequences in scholarly references—page numbers, years, volume/issue numbers, and similar citation digits.

    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
    aurav
    -0.91
     среднего
    -0.85
    terus
    -0.84
    ėk
    -0.83
     részt
    -0.83
    裨
    -0.82
     until
    -0.81
    直到
    -0.78
    居住
    -0.78
     ώ
    -0.77
    POSITIVE LOGITS
     tomba
    1.12
     requerida
    0.99
    两天
    0.96
     rengi
    0.96
    两次
    0.92
     suporta
    0.92
    ῖς
    0.91
     müm
    0.90
    理由は
    0.88
     semplice
    0.87
    Activations Density 0.002%

    No Known Activations