© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 2860
    Prev
    Next
    INDEX
    Explanations

    list `[]` or `()`

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

    The neuron fires on the final content token immediately before the end‐of‐sequence marker.

    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
    ह
    0.82
     კ
    0.82
    4
    0.81
    п
    0.80
    গ
    0.80
    5
    0.79
    6
    0.77
    ─
    0.77
    га
    0.77
    3
    0.77
    POSITIVE LOGITS
     accessor
    0.81
     Dance
    0.76
    COMANDA
    0.75
     kaart
    0.75
     عادت
    0.74
     EXEC
    0.74
     dance
    0.73
     গ্রস্থ
    0.71
     blueprints
    0.70
     Karam
    0.68
    Activations Density 0.072%

    No Known Activations