© 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. 129059
    Prev
    Next
    INDEX
    Explanations

    albums and bands

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

    The neuron is detecting numeric and ordinal tokens (digits and words like “album,” “studio,” “compilation,” “sixth,” “fifteenth,” etc.) that typically appear in music‐related metadata or section headings (e.g. album titles, release years, track listings).

    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
     queridos
    -1.05
     already
    -1.05
     logros
    -1.04
    を見ながら
    -1.04
     fría
    -1.02
    lge
    -1.02
     onOptions
    -1.01
     eşofman
    -1.00
    predicted
    -0.98
     tratado
    -0.98
    POSITIVE LOGITS
    ).
    1.17
     번째
    1.16
    スマホ
    1.16
     then
    1.09
    最初に
    1.07
     their
    1.02
     '');
    1.01
    %%%%%%%%%%%%%%%%
    1.01
     ""),
    1.01
     präsentieren
    1.00
    Activations Density 0.212%

    No Known Activations