© 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. 12544
    Prev
    Next
    INDEX
    Explanations

    Luhansk

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

    The neuron responds strongly to infrequent or unusually specific words—i.e. it flags rare or low-frequency vocabulary items.

    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
     gwiaz
    0.80
     Matcha
    0.79
    折り
    0.78
     み
    0.77
     pequenas
    0.76
     frut
    0.76
    CCOCC
    0.76
     ।”
    0.75
    もちゃ
    0.75
     Tvam
    0.75
    POSITIVE LOGITS
    ра
    0.82
     таки
    0.79
    Acknowledgement
    0.75
    ITH
    0.73
    ivers
    0.71
    sig
    0.70
    竟
    0.70
    sar
    0.69
    곳
    0.68
    sell
    0.68
    Activations Density 0.000%

    No Known Activations