© 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. 33147
    Prev
    Next
    INDEX
    Explanations

    count of amounts or proportions

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

    The neuron fires on numeric literals (integer or decimal numbers) in the text or code.

    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
     filós
    -1.40
     umano
    -1.36
    ystemen
    -1.18
    uransi
    -1.18
    corbic
    -1.15
    Esperamos
    -1.14
     aprendiz
    -1.13
    Pág
    -1.13
    上げます
    -1.10
     стихотво
    -1.10
    POSITIVE LOGITS
     total
    1.38
    −/−
    1.24
    ּוֹ
    1.19
     apresentado
    1.15
    WEBPACK
    1.11
     earlier
    1.10
    hdd
    1.09
    搽
    1.09
    hagic
    1.09
     idyl
    1.07
    Activations Density 0.002%

    No Known Activations