© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 14876
    Prev
    Next
    INDEX
    Explanations

    or followed by define/die/set

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

    The neuron fires on occurrences of boolean/logical operators (e.g. “or”, “&&”, “||” and similar).

    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
     araba
    -1.41
     cilt
    -1.26
     tirs
    -1.14
     igen
    -1.13
     mesures
    -1.11
     uang
    -1.11
     etter
    -1.11
     constituer
    -1.09
     diren
    -1.09
    kten
    -1.08
    POSITIVE LOGITS
     and
    1.32
     but
    1.13
     tendrá
    1.03
    ꈍ
    1.02
     приводит
    1.02
    heria
    1.00
     will
    1.00
     makes
    0.99
    nolia
    0.99
     gän
    0.99
    Activations Density 0.004%

    No Known Activations