© 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. 25391
    Prev
    Next
    INDEX
    Explanations

    Hyundai Motor, Modern Living

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

    The neuron detects tokens that are in all‐caps (e.g. section headings, labels, speaker IDs, or other fully uppercase words).

    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
    〜。
    -0.96
     повіт
    -0.83
    While
    -0.83
    However
    -0.83
     schreibt
    -0.82
    the
    -0.82
     anunciou
    -0.81
     juiz
    -0.81
     RAILROAD
    -0.81
     However
    -0.81
    POSITIVE LOGITS
    ými
    1.16
    ichord
    1.13
    airro
    1.12
    ("-",
    1.09
    billon
    1.09
     funcionalidad
    1.07
     descuentos
    1.06
    Uc
    1.05
    visning
    1.05
     elaboración
    1.05
    Activations Density 0.002%

    No Known Activations