© 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. 15514
    Prev
    Next
    INDEX
    Explanations

    comprar objetos

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

    This neuron is essentially a “junk” detector—it lights up only on extremely rare or anomalous single tokens (e.g. mis-tokenizations or out-of-vocabulary fragments) rather than any linguistic pattern.

    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
     ค
    0.82
    ма
    0.73
     г
    0.73
    。[
    0.72
     =
    0.71
     。
    0.71
    0.71
    ↵↵
    0.70
     б
    0.70
    Ро
    0.69
    POSITIVE LOGITS
    ruits
    0.91
    eted
    0.87
    inating
    0.84
    istoire
    0.82
     मदत
    0.82
     কাটিয়ে
    0.79
    ന്തു
    0.79
    reathe
    0.79
    䅡
    0.79
     climates
    0.78
    Activations Density 0.000%

    No Known Activations