© 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. 14058
    Prev
    Next
    INDEX
    Explanations

    random number generation contexts

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

    This neuron responds most strongly to low‐frequency or otherwise uncommon tokens, effectively flagging rare content words across contexts.

    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
     Consulate
    0.88
    大
    0.80
     salary
    0.79
     Lua
    0.79
     morals
    0.78
     gaji
    0.77
     niece
    0.75
     smuggling
    0.75
     WordPress
    0.74
     Ruby
    0.74
    POSITIVE LOGITS
     impedance
    0.68
    prescription
    0.68
    gten
    0.67
    𝒅
    0.66
    сті
    0.65
    恴
    0.64
    debounce
    0.64
     اكثر
    0.64
    ราง
    0.64
    یم
    0.63
    Activations Density 0.000%

    No Known Activations