© 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. 34-GEMMASCOPE-RES-131K
    7. 64196
    Prev
    Next
    INDEX
    Explanations

    meaning crystal

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

    This neuron primarily responds to high-frequency function words and punctuation (e.g., commas, periods, “in,” “the,” “and”) typical of encyclopedic or Wikipedia‐style text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     enregistrer
    -0.84
     Kró
    -0.82
    мә
    -0.78
    putstr
    -0.77
     pazi
    -0.77
    iedo
    -0.76
     pengiriman
    -0.75
    なりに
    -0.75
    Pics
    -0.75
     jakieś
    -0.75
    POSITIVE LOGITS
     Schott
    0.96
    ーパー
    0.85
    假设
    0.85
     popu
    0.85
     JT
    0.84
    ᐈ
    0.84
     playfully
    0.84
    ={[
    0.83
     podľa
    0.83
     militares
    0.82
    Activations Density 0.003%

    No Known Activations