© 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. 31-GEMMASCOPE-2-RES-16K
    7. 9054
    Prev
    Next
    INDEX
    Explanations

    going beyond, absorbing, or ensuring

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

    The neuron fires most strongly on rare or specialized content words (e.g. long or infrequent proper nouns and technical terms).

    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_31_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
     aromatic
    0.67
    )-
    0.65
     Wirkung
    0.64
     ELISA
    0.62
     sniff
    0.62
     Vicky
    0.62
     poly
    0.61
     fragrant
    0.61
    “-
    0.61
    maxX
    0.61
    POSITIVE LOGITS
    においては
    0.87
    政府
    0.79
    主管
    0.75
     государственных
    0.74
    ધાર
    0.73
     સરકાર
    0.72
     gestão
    0.72
     pemerintah
    0.71
     सरकार
    0.71
     responsabilité
    0.71
    Activations Density 0.000%

    No Known Activations