© 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. 129180
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on numeric expressions (e.g. digits or measurements).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    young people, bigger than, referring to

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    giv
    -0.92
     pria
    -0.91
    timbangan
    -0.90
    zieh
    -0.87
     afast
    -0.86
    éte
    -0.86
    ͟
    -0.85
    criminate
    -0.85
    lapping
    -0.83
    ~”
    -0.81
    POSITIVE LOGITS
    ároz
    0.92
    evre
    0.90
     inocente
    0.89
    ruppen
    0.86
     again
    0.86
    catchError
    0.82
     检查
    0.81
     soaked
    0.78
     hence
    0.78
     obedient
    0.77
    Activations Density 0.093%

    No Known Activations