© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 785
    Prev
    Next
    INDEX
    Explanations

    placement or direction of objects

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

    This neuron fires on the paragraph’s central topic term—i.e. the main noun naming what the section is about.

    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
     ))
    0.46
    
    0.45
     Simulation
    0.43
     Apakah
    0.43
     Synchronization
    0.43
     Communication
    0.42
    artige
    0.41
    )))
    0.41
    Exception
    0.41
     '}
    0.41
    POSITIVE LOGITS
     into
    0.95
    into
    0.72
     إلى
    0.71
     इनटू
    0.65
    到一个
    0.62
     onto
    0.59
     الى
    0.59
     vào
    0.58
     Into
    0.57
     туда
    0.57
    Activations Density 0.000%

    No Known Activations