© 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. 22-GEMMASCOPE-RES-131K
    4. 98513
    Prev
    Next
    INDEX
    Explanations

    programming and machine learning code

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

    The neuron activates on numeric literals (multi‐digit numbers) in the text.

    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
     if
    -1.67
     for
    -1.61
     used
    -1.31
     is
    -1.29
     when
    -1.22
     from
    -1.16
     and
    -1.16
     before
    -1.09
     while
    -1.05
     v
    -1.04
    POSITIVE LOGITS
     restricciones
    1.16
     régal
    1.16
    occurs
    1.14
    蛱
    1.10
    diğiniz
    1.09
    mability
    1.09
     fonctionnent
    1.08
     teorías
    1.08
     mortes
    1.05
    ïa
    1.05
    Activations Density 0.012%

    No Known Activations