© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 22141
    Prev
    Next
    INDEX
    Explanations

    Portuguese questions about functionality

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

    The neuron responds to Portuguese question words—particularly “como” and “que”—marking the presence of an interrogative.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    l
    -2.91
    );
    -2.38
     of
    -2.30
    }
    -2.30
     for
    -2.14
     so
    -2.09
     Hinblick
    -2.02
    -
    -2.00
     strict
    -1.98
    ↵
    -1.97
    POSITIVE LOGITS
    饜
    2.55
     expédi
    2.30
     seleccionadas
    2.23
    stood
    2.20
    腚
    2.19
     抠
    2.16
    cimentos
    2.14
     graisse
    2.13
     vulné
    2.11
     refroidissement
    2.09
    Activations Density 0.003%

    No Known Activations