© 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
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 3514
    Prev
    Next
    INDEX
    Explanations

    code initialization for `x` and `y` coordinates

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

    The neuron fires on rare or out‐of‐vocabulary tokens (e.g. proper names, acronyms, odd punctuated strings), effectively flagging unusual or domain‐specific words.

    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
    部分
    0.80
     introns
    0.77
     setae
    0.75
    нский
    0.75
     lysis
    0.70
    એ
    0.70
     Resolve
    0.69
     listar
    0.69
     نفت
    0.69
     srcs
    0.69
    POSITIVE LOGITS
    fontsize
    0.81
     चुप
    0.79
     Beaches
    0.77
     iż
    0.76
    dough
    0.74
    comedy
    0.74
    CONSUM
    0.73
    handsome
    0.73
    씌
    0.73
    diamond
    0.71
    Activations Density 0.000%

    No Known Activations