© 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. 3209
    Prev
    Next
    INDEX
    Explanations

    and describe or implement

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

    The neuron strongly activates on high‐impact scientific contribution verbs (e.g. “generate,” “demonstrated,” “provides,” “implemented”) that describe core research actions and results.

    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
     ;\
    0.76
    ;",
    0.70
     olmuş
    0.67
     perny
    0.64
     nitrog
    0.64
     -\
    0.63
    ;\
    0.63
     samoglas
    0.62
     haberse
    0.62
     squamous
    0.61
    POSITIVE LOGITS
    rology
    0.64
    <unused2100>
    0.61
    ERSON
    0.58
    ulkan
    0.55
    落实
    0.55
    ாடி
    0.55
    হাব
    0.54
     entreten
    0.54
    移行
    0.54
    ूद
    0.53
    Activations Density 0.439%

    No Known Activations