© 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. 24-GEMMASCOPE-2-RES-16K
    7. 14933
    Prev
    Next
    INDEX
    Explanations

    giữa demo và bảo hiểm

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

    The neuron strongly activates on verbs that express perception, cognition, or affect (e.g. understand, feel, wonder) often used in visitor-experience or user-experience statements.

    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
    (
    1.12
    2
    1.09
    1
    1.07
    or
    1.02
    /
    1.01
    6
    0.98
    )
    0.97
    8
    0.95
    and
    0.95
    ites
    0.95
    POSITIVE LOGITS
     nhiều
    1.82
     như
    1.72
     nhìn
    1.67
     những
    1.63
     thời
    1.63
     nhưng
    1.61
     phát
    1.59
     giúp
    1.59
     khi
    1.58
     hoàn
    1.58
    Activations Density 0.000%

    No Known Activations