© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 7955
    Prev
    Next
    INDEX
    Explanations

    cumulative count, one part

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

    The neuron fires strongly on content-bearing (i.e. relatively rare or semantically informative) words rather than on common function 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
     เงี้ย
    0.88
    ￣
    0.79
    стым
    0.79
     популација
    0.78
     Naina
    0.78
     Oké
    0.76
    UAGE
    0.75
    дных
    0.74
    як
    0.73
    頎
    0.73
    POSITIVE LOGITS
     від
    0.84
    其实
    0.81
    सुद्धा
    0.75
     frigor
    0.74
     fratello
    0.73
    ğinde
    0.73
     hơi
    0.72
    thed
    0.71
    গ
    0.71
     спів
    0.70
    Activations Density 0.000%

    No Known Activations