© 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. 14474
    Prev
    Next
    INDEX
    Explanations

    explains novel report online framework

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

    This neuron doesn’t actually pick out any coherent linguistic pattern – its “activations” are just spurious, high‐magnitude artifacts rather than any meaningful token class.

    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
     samano
    0.91
    𝒆
    0.86
     গবে
    0.84
    𝒂
    0.82
    ărilor
    0.80
    𝒕
    0.79
    ا۔
    0.78
    hardness
    0.78
     oreille
    0.77
    𝒐
    0.76
    POSITIVE LOGITS
    績
    0.69
     sculpted
    0.64
     Nä
    0.64
    الصفحه
    0.64
     person
    0.63
     unpacked
    0.63
    </strong>
    0.62
     Painted
    0.62
     wach
    0.62
     Regents
    0.61
    Activations Density 0.001%

    No Known Activations