© 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. 31-GEMMASCOPE-2-RES-16K
    7. 3125
    Prev
    Next
    INDEX
    Explanations

    numbers and short words

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

    The neuron fires on key action or transaction words (e.g. contribute, transfer, click, owned).

    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
    rü
    0.51
    gulp
    0.48
     bones
    0.47
     …
    0.46
    sa
    0.46
    nu
    0.45
    <em>
    0.44
    ump
    0.44
    essions
    0.44
    ayed
    0.44
    POSITIVE LOGITS
     같
    0.54
    같
    0.52
    সহিত
    0.51
     prothorax
    0.51
     einfachen
    0.49
    딴
    0.48
     Prothorax
    0.48
    쨰
    0.48
    佢
    0.48
    Kemudian
    0.48
    Activations Density 0.006%

    No Known Activations