© 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. 31-GEMMASCOPE-2-RES-16K
    7. 11155
    Prev
    Next
    INDEX
    Explanations

    specific tokens after punctuation

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

    The neuron activates on prominent “news”-style headline words—especially terms like “news,” “Headlines,” “Breaking,” and other high‐visibility title tokens.

    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_31_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
    haltung
    0.62
    ）』
    0.54
     جام
    0.50
     大き
    0.50
    wale
    0.50
    நாட்டு
    0.50
    斗
    0.50
    gamb
    0.50
     多い
    0.50
     vu
    0.50
    POSITIVE LOGITS
     Pá
    0.55
     earmarked
    0.55
     egyes
    0.52
     subsequ
    0.52
     инг
    0.50
     exacerbated
    0.49
     subheading
    0.48
    áticamente
    0.47
    POSED
    0.47
    }}{\
    0.47
    Activations Density 0.148%

    No Known Activations