© 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. 1903
    Prev
    Next
    INDEX
    Explanations

    identifying actions or processes

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

    The neuron is sensitive to words ending in “-ing,” i.e. gerunds or present-participle verb forms.

    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
    ‌ی
    0.71
    siege
    0.64
    Assalamualaikum
    0.64
    ן
    0.63
     aperta
    0.62
    يل
    0.61
    ‌ای
    0.61
    が付
    0.60
    ى
    0.59
    <0xD3>
    0.59
    POSITIVE LOGITS
    ع
    0.84
    redients
    0.75
    ly
    0.73
    他
    0.73
    the
    0.72
    gom
    0.66
    地
    0.66
    过程
    0.66
    주
    0.65
    된
    0.65
    Activations Density 0.000%

    No Known Activations