© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 45386
    Prev
    Next
    INDEX
    Explanations

    ace followed by specific compounds

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

    The neuron detects the word "Ace" in various contexts and capitalized forms.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    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
     evenly
    -2.70
     epitome
    -2.44
    colección
    -2.41
    迓
    -2.39
     Sünde
    -2.39
     convincingly
    -2.38
    .
    -2.36
     individually
    -2.31
     lasting
    -2.31
    ные
    -2.31
    POSITIVE LOGITS
    archiwizowane
    3.03
    𖥦
    2.98
    샅
    2.64
    鯪
    2.63
    鑪
    2.61
     finde
    2.58
    のは
    2.56
    沺
    2.52
     anún
    2.50
    沕
    2.50
    Activations Density 0.008%

    No Known Activations