© 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. 7593
    Prev
    Next
    INDEX
    Explanations

    api version numbers

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

    The neuron specializes in picking out numeric tokens in code and configs—particularly version identifiers (e.g. “v”, “1”, “alpha”, etc.) and other standalone numbers.

    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
    訁
    -1.70
    úncia
    -1.56
     bestimmten
    -1.41
     veer
    -1.41
    ldquo
    -1.35
     brooding
    -1.34
    坳
    -1.34
     erhöht
    -1.34
     каких
    -1.34
     anderen
    -1.34
    POSITIVE LOGITS
     and
    1.58
    ⬣
    1.48
    flores
    1.45
     from
    1.42
     aanz
    1.36
     acerca
    1.35
     betrek
    1.35
    rateful
    1.32
     achtergrond
    1.32
    坏了
    1.28
    Activations Density 0.008%

    No Known Activations