© 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. 1549
    Prev
    Next
    INDEX
    Explanations

    reference list entries

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

    The neuron selectively activates on words that appear as key title or heading tokens in academic paper references (i.e. the main capitalized title words of scholarly articles).

    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
    .??.??"]
    0.48
     kttt
    0.46
    jTextField
    0.46
    HWND
    0.45
    UnderWater
    0.44
     अपॉन
    0.44
     Оюн
    0.44
     haught
    0.44
    Nếu
    0.44
     पद्
    0.43
    POSITIVE LOGITS
     and
    0.61
     four
    0.51
     for
    0.49
     K
    0.48
     D
    0.47
     from
    0.47
     désormais
    0.47
     A
    0.46
     categories
    0.46
     P
    0.46
    Activations Density 0.018%

    No Known Activations