© 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. 95538
    Prev
    Next
    INDEX
    Explanations

    questions or statements

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

    The neuron detects hexadecimal digits (e.g. characters like “8”) in code or address literals.

    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
     technik
    -1.02
     kellett
    -0.91
     PUSTAKA
    -0.89
     informaci
    -0.87
     joli
    -0.86
    ections
    -0.85
     servir
    -0.83
    Затем
    -0.83
     rosas
    -0.83
     botas
    -0.83
    POSITIVE LOGITS
     tatsächlich
    1.03
    Algérie
    0.98
     empfohlen
    0.93
     diferite
    0.92
     первым
    0.89
     instead
    0.89
     acercar
    0.88
     Apalagi
    0.88
     vielmehr
    0.88
    IAB
    0.88
    Activations Density 0.000%

    No Known Activations