© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 15567
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on low-frequency (rare) words—i.e. uncommon tokens in the text.

    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
     gminy
    0.85
     哈
    0.80
     ></
    0.68
    Примеча
    0.68
     polož
    0.67
     Ừ
    0.66
     Gecko
    0.65
    $_
    0.63
     Waalaikumsalam
    0.63
     Hei
    0.63
    POSITIVE LOGITS
     associates
    0.75
    backpack
    0.73
     commits
    0.71
    وْ
    0.71
    commitment
    0.71
    обра
    0.70
    वर
    0.70
     maquillaje
    0.69
     cumulative
    0.69
     intérêt
    0.69
    Activations Density 0.000%

    No Known Activations