© 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. 12989
    Prev
    Next
    INDEX
    Explanations

    code structures and array manipulation

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

    The neuron is effectively “dead” — it never activates on any tokens.

    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
     vuotta
    -0.98
    camente
    -0.92
     hiyo
    -0.90
     of
    -0.88
     atât
    -0.85
    不高
    -0.85
    などが
    -0.85
     hvid
    -0.84
     taal
    -0.82
     découver
    -0.82
    POSITIVE LOGITS
     Geduld
    0.90
    );"
    0.85
    unno
    0.79
     Pixel
    0.79
     Job
    0.79
    providedIn
    0.78
     compare
    0.77
    organiser
    0.77
     który
    0.77
    числе
    0.77
    Activations Density 0.002%

    No Known Activations