© 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. 25906
    Prev
    Next
    INDEX
    Explanations

    descriptive noun phrases

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

    The neuron activates on non‐English (particularly Portuguese) words or morphemes in the 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_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
    はずです
    -1.17
     intellectuelle
    -0.82
     normaux
    -0.81
    ären
    -0.81
    スズキ
    -0.80
     teleport
    -0.80
    戔
    -0.79
    ようです
    -0.78
    hamilton
    -0.77
     Prado
    -0.77
    POSITIVE LOGITS
    ynarod
    0.88
     onLoad
    0.85
     provocando
    0.83
     causando
    0.76
     you
    0.76
    agissait
    0.75
     충
    0.75
     TOPOLOGY
    0.73
     unico
    0.73
     Baumwolle
    0.72
    Activations Density 0.007%

    No Known Activations