© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 649
    Prev
    Next
    INDEX
    Explanations

    key part of

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

    The neuron is looking for “of-” partitive constructions—phrases like “part of an effort,” “much of X,” “some of,” etc.

    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_31_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
    0.71
    x
    0.59
     mussels
    0.58
     niektórych
    0.55
    1
    0.55
     제외
    0.55
     solcher
    0.54
     puntos
    0.54
     নির্দিষ্ট
    0.54
    R
    0.54
    POSITIVE LOGITS
     shaping
    0.91
     furthering
    0.90
     society
    0.89
     policym
    0.88
     kehidupan
    0.87
     discussions
    0.83
     educating
    0.81
    推动
    0.80
     deliberations
    0.79
     academia
    0.78
    Activations Density 0.773%

    No Known Activations