© 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. 24-GEMMASCOPE-2-RES-16K
    7. 13497
    Prev
    Next
    INDEX
    Explanations

    explaining or stating something

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

    The neuron fires strongly on reporting or attribution verbs (e.g. "said," "commended," "noted") that introduce quoted or reported speech.

    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
    imah
    0.63
     বলছেন
    0.63
     soprattutto
    0.62
    jResponse
    0.61
    wari
    0.61
     souligne
    0.59
     advices
    0.59
    ⸨
    0.59
     cautions
    0.58
     கொண்டுள்ளது
    0.57
    POSITIVE LOGITS
     that
    0.69
    ō
    0.67
     Somos
    0.64
    ű
    0.61
     huyền
    0.61
    ál
    0.61
     เรา
    0.60
     اینکه
    0.59
    ),]),
    0.59
    ulları
    0.58
    Activations Density 0.007%

    No Known Activations