© 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. 11648
    Prev
    Next
    INDEX
    Explanations

    This neuron consistently spikes on the very first substantial content word at the start of a new paragraph or section.

    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
     soát
    0.96
     Customizable
    0.80
    valho
    0.79
     Ҳ
    0.79
     Tâm
    0.79
     misdeme
    0.75
    然而
    0.75
    χύ
    0.75
     لیکن
    0.74
     সম্মুখ
    0.74
    POSITIVE LOGITS
    Ĭ
    0.78
    HashSet
    0.77
    Ladies
    0.75
    CA
    0.75
    FC
    0.75
    CN
    0.70
    AGRIC
    0.70
    
    0.70
    Goed
    0.69
    Nel
    0.67
    Activations Density 0.000%

    No Known Activations