© 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. 65508
    Prev
    Next
    INDEX
    Explanations

    numbers, dates, and specific names

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

    The neuron fires on program‐oriented tokens (package names, command-line flags, code identifiers, CSS selectors, etc.), i.e. bits of code or config syntax.

    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
    TokenNameDOT
    -0.87
     например
    -0.86
    prior
    -0.86
    OSPITAL
    -0.85
     Section
    -0.82
    reports
    -0.82
     Chapter
    -0.81
     Villa
    -0.80
    section
    -0.80
    control
    -0.79
    POSITIVE LOGITS
    uestos
    0.89
     Liens
    0.87
     روزانه
    0.84
    oys
    0.83
     infrequently
    0.83
    以上
    0.81
     можна
    0.81
     lastly
    0.81
     dernière
    0.80
    Lastly
    0.79
    Activations Density 0.002%

    No Known Activations