© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 6072
    Prev
    Next
    INDEX
    Explanations

    ": " version "

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

    The neuron is picking up numeric tokens with decimal points (i.e. floating‐point numbers).

    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
     a
    0.55
    the
    0.55
    वा
    0.53
    Feedback
    0.53
     자신이
    0.52
    가
    0.51
    пуска
    0.51
    ったり
    0.51
    したり
    0.50
    Disabled
    0.49
    POSITIVE LOGITS
    ։
    0.92
     sejak
    0.79
    .​​
    0.79
    .“
    0.78
    ។
    0.75
     poiché
    0.75
     bởi
    0.73
    .«
    0.71
    .¹
    0.70
    .。
    0.70
    Activations Density 0.000%

    No Known Activations