© 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. 3374
    Prev
    Next
    INDEX
    Explanations

    specifying host or address

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

    The neuron fires strongly on mention of headline-style or “highlight” tokens—proper names, numerical facts (years, ordinals), or extreme/superlative descriptors—i.e. the key entities and figures that news‐flash or headline text would emphasize.

    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
     poskyt
    1.02
    茟
    0.94
    För
    0.90
     действи
    0.88
     gebruiken
    0.88
     práce
    0.86
     werken
    0.84
     lancio
    0.84
     arbet
    0.84
     tersedia
    0.83
    POSITIVE LOGITS
    ك
    0.91
    ,
    0.75
     turnaround
    0.70
    های
    0.66
     controversial
    0.64
     animated
    0.64
     accusation
    0.63
    う
    0.63
     amendments
    0.62
     towering
    0.62
    Activations Density 0.001%

    No Known Activations