© 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. 6271
    Prev
    Next
    INDEX
    Explanations

    bolded text separator

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

    The neuron activates on numeric tokens—especially decimal numbers—flagging occurrences of quantitative data.

    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
    élim
    0.84
    á
    0.77
    é
    0.76
    élo
    0.75
    Ét
    0.75
    ých
    0.75
    ité
    0.74
    après
    0.74
    éz
    0.73
    abá
    0.72
    POSITIVE LOGITS
     AppComponent
    0.82
     Kollegen
    0.82
     HelloWorld
    0.82
     instantiate
    0.78
     P
    0.78
     hoge
    0.78
     J
    0.77
     R
    0.77
     Arduino
    0.77
     spezielle
    0.76
    Activations Density 3.874%

    No Known Activations