© 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. 34-GEMMASCOPE-RES-131K
    4. 107402
    Prev
    Next
    INDEX
    Explanations

    license and software

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

    The neuron strongly activates on numeric tokens—both digit literals and spelled-out numbers.

    oai_token-act-pair · o4-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
    vine
    -0.71
    artz
    -0.68
    utnik
    -0.67
     fullname
    -0.67
     proyección
    -0.66
    onez
    -0.65
     here
    -0.65
     values
    -0.64
     reticulum
    -0.63
     prioridad
    -0.63
    POSITIVE LOGITS
     hubs
    0.76
     Colby
    0.75
     CREATE
    0.71
    gcd
    0.69
    msa
    0.68
     Div
    0.68
     Gin
    0.66
    firi
    0.65
     Choice
    0.65
     Dom
    0.65
    Activations Density 0.061%

    No Known Activations