© 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. 2714
    Prev
    Next
    INDEX
    Explanations

    speaking, questioning, addressing

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

    This neuron lights up on blocks of narrative action or stage-directions (prose describing characters’ movements or setting), rather than on dialogue.

    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
    ↵↵
    0.52
    i
    0.50
    0
    0.46
    ↵
    0.46
    l
    0.46
    P
    0.46
     ).
    0.45
    5
    0.45
    endal
    0.44
     Mỗi
    0.44
    POSITIVE LOGITS
     "\\
    0.64
     Antworten
    0.61
     थोड़ा
    0.60
    ഔ
    0.60
     "?"]
    0.60
     bertanya
    0.59
     erklären
    0.59
     pergunt
    0.57
     первым
    0.57
     \""
    0.57
    Activations Density 0.006%

    No Known Activations