© 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. 72632
    Prev
    Next
    INDEX
    Explanations

    names and company names

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

    The neuron strongly fires on code-specific symbols and numeric literals (operators, punctuation, and numbers) rather than on natural‐language 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
     metter
    -0.82
     empfehlen
    -0.80
     baris
    -0.80
     åt
    -0.78
     from
    -0.76
     Byrne
    -0.76
    gruss
    -0.76
    ственные
    -0.74
    Unsere
    -0.73
    ator
    -0.73
    POSITIVE LOGITS
    return
    0.93
     but
    0.87
    ittin
    0.86
     Unternehmens
    0.81
    iembre
    0.81
     jezelf
    0.81
    cellent
    0.80
    bihan
    0.80
    дента
    0.79
     dagelijks
    0.79
    Activations Density 0.002%

    No Known Activations