© 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. 15133
    Prev
    Next
    INDEX
    Explanations

    OpenAI, GitHub, learning, food, context

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

    The neuron spots technical-style tokens—especially numerals and engineering or scientific terms (e.g. model names, measurements, class or figure numbers).

    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
     ढंग
    0.66
    धीश
    0.61
    ობს
    0.59
    enten
    0.59
    ђено
    0.59
    ட்சத்திர
    0.58
    ريقة
    0.57
    ስቃሴ
    0.57
    的情况下
    0.57
    sorting
    0.56
    POSITIVE LOGITS
     were
    0.78
     should
    0.77
     are
    0.77
     отличаются
    0.76
     aren
    0.72
     jsou
    0.69
    개가
    0.69
     weren
    0.66
     будут
    0.66
     були
    0.65
    Activations Density 0.346%

    No Known Activations