© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 12724
    Prev
    Next
    INDEX
    Explanations

    division, strengths, what

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

    The neuron fires on words that serve as block or sub‐section headings (i.e. title‐case or stand-alone header tokens at the start of new segments).

    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
    ൾ
    0.82
    duh
    0.81
     pró
    0.79
    apar
    0.79
    thel
    0.79
    denum
    0.78
     etn
    0.77
    thal
    0.76
    hattim
    0.75
     tanta
    0.75
    POSITIVE LOGITS
     deployments
    0.82
     instantiation
    0.78
     absences
    0.73
    碉
    0.73
     পারেনা
    0.73
     Ще
    0.73
     coordinate
    0.71
     Upload
    0.71
     TLabel
    0.71
    存档
    0.70
    Activations Density 0.001%

    No Known Activations