© 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. 8736
    Prev
    Next
    INDEX
    Explanations

    Configured :**

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

    The neuron fires on the single most content‐bearing word in each passage—i.e. the key noun or adjective that carries the main topic or gist.

    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
     whe
    0.82
    ెల
    0.81
     kolay
    0.80
     ನ
    0.80
    ពេល
    0.79
     కంటే
    0.78
     ਪ
    0.77
    сть
    0.77
     फॉर्मेट
    0.77
    서
    0.77
    POSITIVE LOGITS
    speech
    0.78
               
    0.77
     Dubrov
    0.74
    <unused60>
    0.73
    structural
    0.72
             
    0.71
    sphere
    0.71
    subtract
    0.70
    commun
    0.70
    “.
    0.70
    Activations Density 0.000%

    No Known Activations