© 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. 15343
    Prev
    Next
    INDEX
    Explanations

    call to action

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

    The neuron strongly activates on Spanish-language tokens (e.g. “Qué,” “te,” “inspira,” “de,” “empresa,” etc.).

    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
     hjemmeside
    -1.01
    digitalWrite
    -1.00
    Ucraina
    -0.98
     bereits
    -0.97
     ホイール
    -0.93
     titular
    -0.92
    лити
    -0.91
     pamamagitan
    -0.90
     keramik
    -0.88
    tür
    -0.87
    POSITIVE LOGITS
     jakieś
    1.07
    ("}");
    1.07
    Aggi
    1.05
     dodatkowe
    0.99
    Were
    0.98
    <0xF3>
    0.97
     erwor
    0.97
    сында
    0.96
    了口气
    0.94
    𝔨
    0.94
    Activations Density 0.128%

    No Known Activations