© 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. 24-GEMMASCOPE-2-RES-16K
    7. 13493
    Prev
    Next
    INDEX
    Explanations

    left/right/0 movement

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

    The neuron fires strongly on standalone substantive nouns or labels—especially topic/category terms (e.g. “relationships,” “Epistle,” “treatment,” “marketing,” “science,” “PS”) that often head lists or denote key sections.

    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_24_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
    ೀ
    0.79
    ιν
    0.76
    ayos
    0.73
    蜥
    0.73
     terrorist
    0.71
    岸
    0.71
    ।
    0.70
    াসের
    0.70
    az
    0.69
    ’
    0.68
    POSITIVE LOGITS
     assim
    0.83
     fiche
    0.82
     transp
    0.81
     assinatura
    0.77
     imprim
    0.76
     trovare
    0.76
     onde
    0.74
    क्षक
    0.73
     pâ
    0.73
     findest
    0.73
    Activations Density 0.000%

    No Known Activations