© 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. 22-GEMMASCOPE-RES-131K
    7. 33767
    Prev
    Next
    INDEX
    Explanations

    mathematical expressions including variables

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

    The neuron strongly activates on numeric literals (digits, integers, and decimals) appearing 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_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
     海报
    -0.82
    サル
    -0.80
     Hog
    -0.79
     hỗ
    -0.77
     encara
    -0.75
    いただきました
    -0.73
     like
    -0.72
    うつ
    -0.72
    幾個
    -0.72
     bilgiler
    -0.72
    POSITIVE LOGITS
     biografi
    0.91
     Cadena
    0.84
     metabolic
    0.82
     refunded
    0.82
     Lutheran
    0.81
    езд
    0.80
     construc
    0.80
    моно
    0.79
     掛け
    0.79
     gué
    0.77
    Activations Density 0.077%

    No Known Activations