© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 9601
    Prev
    Next
    INDEX
    Explanations

    Yeah, okay, so

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

    This neuron fires on lyric‐style vocalizations or melodic syllables (e.g. “la,” “hey-yo,” repeated “la-la” patterns) found in song lyrics.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ’.
    0.64
     Hogwarts
    0.64
    ပါသည်။
    0.64
    .’”
    0.63
     .”
    0.62
     posterity
    0.61
     MOST
    0.61
     ❤️
    0.61
    .’
    0.60
     🥰
    0.60
    POSITIVE LOGITS
     yeah
    0.77
     Yeah
    0.68
     gonna
    0.67
    yeah
    0.66
    Yeah
    0.64
    睜
    0.62
    演唱
    0.61
    んで
    0.60
    য়াল
    0.59
     lyrics
    0.59
    Activations Density 0.072%

    No Known Activations