© 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. 26420
    Prev
    Next
    INDEX
    Explanations

    numbers and words

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

    The neuron is keyed to spotting floating-point numeric literals—i.e. numbers containing a decimal point.

    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
     to
    -1.10
     предлагает
    -0.88
    viaf
    -0.88
     Begleitung
    -0.88
    ᄃ
    -0.85
    できるので
    -0.84
    せずに
    -0.83
     membuat
    -0.82
    反应过来
    -0.82
    znak
    -0.80
    POSITIVE LOGITS
     Synag
    0.91
     baden
    0.89
    ítő
    0.88
     zwi
    0.83
     Konkur
    0.83
    uminous
    0.83
     wies
    0.82
    亡
    0.82
     Nguy
    0.82
    biology
    0.82
    Activations Density 0.160%

    No Known Activations