© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 95630
    Prev
    Next
    INDEX
    Explanations

    items and vocabulary

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

    The neuron is specialized in spotting numeric tokens and measurement/specification values (e.g. digits, units, and technical counts).

    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
     on
    -0.97
    </strong>
    -0.91
    </em>
    -0.88
    场合
    -0.83
     notwendig
    -0.81
    Yogurt
    -0.79
    のだろう
    -0.79
     to
    -0.77
     can
    -0.77
    ᧐
    -0.76
    POSITIVE LOGITS
     camiset
    1.07
    нили
    1.00
     toalha
    0.96
     appartiennent
    0.93
    ressed
    0.93
     perten
    0.92
    bsen
    0.92
     ᾿
    0.92
     DataAccess
    0.92
     alça
    0.91
    Activations Density 0.017%

    No Known Activations