© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 20067
    Prev
    Next
    INDEX
    Explanations

    but/mas/но/aber/yet

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

    This neuron fires on adversative conjunctions (“but” equivalents) in multiple languages.

    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
    <td>
    -2.72
     of
    -2.63
     für
    -2.39
    镡
    -2.34
    
    -2.33
    ⼥
    -2.31
    踅
    -2.23
    谖
    -2.22
     cuz
    -2.22
    на
    -2.19
    POSITIVE LOGITS
     evitando
    1.89
    véd
    1.88
    {
    1.88
     行き
    1.88
     Automó
    1.85
     blauen
    1.83
     dunklen
    1.82
    レストラン
    1.82
    quatre
    1.79
     mesmos
    1.78
    Activations Density 0.005%

    No Known Activations