© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 79327
    Prev
    Next
    INDEX
    Explanations

    names and foreign words

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

    The neuron activates on standalone numeric tokens (e.g. multi‐digit numbers, especially four‐digit years).

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    irling
    -0.78
    üglich
    -0.72
    当時
    -0.70
     później
    -0.70
     médié
    -0.69
     результат
    -0.69
    Cher
    -0.69
    Dig
    -0.68
     człowieka
    -0.68
     Otro
    -0.67
    POSITIVE LOGITS
     doom
    0.81
     dynasty
    0.77
     continuo
    0.73
     implementing
    0.72
     gom
    0.70
    纵
    0.70
    CWE
    0.69
     jsPsych
    0.69
    PTP
    0.68
     осно
    0.68
    Activations Density 0.019%

    No Known Activations