© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 44660
    Prev
    Next
    INDEX
    Explanations

    next term

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

    The neuron chiefly activates on numeric tokens and numerical expressions (i.e., any digits or numbers in the 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_10/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
     Municipalidad
    -1.55
    という
    -1.49
    見て
    -1.48
    Comes
    -1.48
    have
    -1.48
    찮
    -1.42
     зимние
    -1.41
     совершенно
    -1.41
    のは
    -1.41
     pue
    -1.40
    POSITIVE LOGITS
    翆
    2.30
    ፨
    1.77
    ࣭
    1.73
    甚至是
    1.68
    ࿊
    1.67
    涠
    1.65
    偸
    1.63
    ৄ
    1.63
    玊
    1.63
     copious
    1.56
    Activations Density 0.028%

    No Known Activations