© 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. 65015
    Prev
    Next
    INDEX
    Explanations

    identification tags

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

    The neuron activates almost exclusively on numeric tokens, i.e. any standalone number.

    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
    ాన
    -1.00
    gangenen
    -0.98
     sema
    -0.96
     voto
    -0.95
     Filt
    -0.95
     DATABASE
    -0.94
    ⩽
    -0.94
    ➕
    -0.94
    $-\
    -0.94
    殃
    -0.94
    POSITIVE LOGITS
     both
    0.86
     how
    0.82
    jooq
    0.82
     Wahr
    0.81
    あたり
    0.80
    까
    0.76
     most
    0.76
     eff
    0.76
     M
    0.75
     powied
    0.75
    Activations Density 0.000%

    No Known Activations