© 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. 13160
    Prev
    Next
    INDEX
    Explanations

    perception, reality, home, tied

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

    This neuron fires on the single most salient content word in each passage—the key topic or “headline” word around which the paragraph centers.

    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
    おう
    0.80
    )
    0.77
    ()
    0.71
    callbacks
    0.71
    作業
    0.69
    جاه
    0.68
    燗
    0.67
    થી
    0.66
     Algebraic
    0.65
     Astrology
    0.65
    POSITIVE LOGITS
     moulded
    0.82
     annexed
    0.80
    निमल
    0.79
     excelled
    0.76
    ﮏ
    0.76
     steeply
    0.75
     rejuvenated
    0.75
    İZ
    0.75
    ﻊ
    0.71
     incorporated
    0.70
    Activations Density 0.000%

    No Known Activations