© 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. 9835
    Prev
    Next
    INDEX
    Explanations

    words ending in -ip, -op, -ipster, etc

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

    The neuron is primarily triggered by numeric measurements or statistical figures (especially floating-point numbers) in the text.

    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
    厶
    0.79
    ل
    0.71
    تس
    0.64
    ج
    0.63
    ্র
    0.63
    दछ
    0.61
     disheart
    0.61
    깐
    0.61
     мод
    0.60
    pump
    0.60
    POSITIVE LOGITS
    ocalypse
    1.06
    munk
    1.00
    sters
    0.98
    wreck
    0.86
    니다
    0.82
    ockets
    0.80
    sticks
    0.79
    nosis
    0.77
    sey
    0.76
    shots
    0.75
    Activations Density 0.722%

    No Known Activations