© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 8246
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on any token that begins a new line or paragraph—i.e. the first word of a freshly started sentence/segment.

    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
    ∈
    0.82
    }+
    0.79
    }$\
    0.75
    ৩৭
    0.71
    ﺮ
    0.70
     Shil
    0.68
    <0xBE>
    0.67
    ৪২
    0.67
     कैरेट
    0.67
    鄀
    0.67
    POSITIVE LOGITS
    aneity
    0.77
     aidant
    0.77
     ይታ
    0.75
    viron
    0.74
    irà
    0.74
    uidos
    0.73
    annuation
    0.73
    ücksicht
    0.73
     équipements
    0.71
    éis
    0.71
    Activations Density 0.001%

    No Known Activations