© 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. 3883
    Prev
    Next
    INDEX
    Explanations

    pipeline usage and text processing

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

    The neuron fires on central topic words—key nouns or other content words that signal the main subject of a passage.

    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
    žad
    0.83
    że
    0.79
    threatening
    0.78
    friend
    0.76
    з
    0.76
    কার
    0.74
    ếu
    0.74
    şim
    0.74
     йөк
    0.73
    emorrh
    0.73
    POSITIVE LOGITS
     hadrons
    0.85
    ع
    0.83
     goggles
    0.82
    ต่างๆ
    0.82
     APIs
    0.80
     JSONArray
    0.78
     headsets
    0.73
     TECHNIQUES
    0.73
     nums
    0.73
     KRS
    0.73
    Activations Density 0.000%

    No Known Activations