© 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
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 10668
    Prev
    Next
    INDEX
    Explanations

    be restricted to

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

    The neuron strongly activates on tokens that occur very far along (at high absolute positions) in a long input sequence.

    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
    :
    0.82
    ;
    0.77
    >
    0.77
    ।
    0.76
    گ
    0.76
    .
    0.75
    ка
    0.74
    {
    0.73
     замести
    0.71
    وجه
    0.70
    POSITIVE LOGITS
     decisions
    0.83
    ្វី
    0.78
     antitrust
    0.76
     agglomer
    0.75
     monopolies
    0.74
    spiel
    0.74
    統合
    0.73
     groves
    0.72
     निर्धारण
    0.71
     бизне
    0.71
    Activations Density 0.000%

    No Known Activations