© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 109269
    Prev
    Next
    INDEX
    Explanations

    Ka followed by syllables/words

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

    This neuron activates strongly on the specific two‐letter token “KA” (in all‐caps), i.e. the standalone bigram “KA.”

    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
     inconveniente
    -0.91
     Deta
    -0.85
     föruts
    -0.83
    と思いました
    -0.81
     lacri
    -0.80
    回戦
    -0.79
     ヴィトン
    -0.77
     Toxicity
    -0.76
    知り
    -0.75
     sini
    -0.74
    POSITIVE LOGITS
     Ka
    1.09
     ka
    0.95
    Ka
    0.94
     absolutamente
    0.88
    Thankfully
    0.83
     entraîner
    0.82
    的一切
    0.82
     KA
    0.82
    interpr
    0.81
     involucra
    0.79
    Activations Density 0.019%

    No Known Activations