© 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. 1183
    Prev
    Next
    INDEX
    Explanations

    engine, load, error, simpler

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

    The neuron fires strongly on the first substantive word of each new text segment or paragraph.

    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
     Amalfi
    0.83
     Kavanaugh
    0.75
     Puglia
    0.75
    ﻜ
    0.75
    縄
    0.74
     Asha
    0.73
     HIM
    0.73
    ؎
    0.72
    ༨
    0.71
    פה
    0.71
    POSITIVE LOGITS
    ose
    0.81
    apd
    0.75
    듕
    0.74
    ots
    0.72
    ais
    0.71
    zonych
    0.71
     વિચાર
    0.71
    ost
    0.69
    ients
    0.68
    osition
    0.66
    Activations Density 0.004%

    No Known Activations