© 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. 22314
    Prev
    Next
    INDEX
    Explanations

    almost daily, so that, in unintended

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

    The neuron fires on numerical tokens (both integers and decimals).

    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
     its
    -2.08
     whose
    -1.87
     his
    -1.73
     if
    -1.68
     in
    -1.63
     these
    -1.54
     about
    -1.53
     my
    -1.52
     what
    -1.48
     it
    -1.41
    POSITIVE LOGITS
    Kategorie
    1.74
    你不是
    1.41
    inaison
    1.41
    又是一
    1.39
     végétal
    1.38
    Reklama
    1.36
    Hvordan
    1.33
    子供が
    1.31
    這邊
    1.31
    ブログ村
    1.31
    Activations Density 0.305%

    No Known Activations