© 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. 2957
    Prev
    Next
    INDEX
    Explanations

    The neuron consistently spikes on the first substantial content word at the start of a new segment or paragraph.

    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
     appellants
    0.94
    委会
    0.83
    inė
    0.83
     coffers
    0.83
    फोर्स
    0.82
    ᱖
    0.81
    承包人
    0.81
     dissidents
    0.80
    ค
    0.80
    ୫
    0.80
    POSITIVE LOGITS
    aside
    0.70
     Dot
    0.64
     যাবত
    0.64
    мое
    0.63
     varit
    0.63
    жден
    0.62
     проходи
    0.59
    견
    0.58
     щ
    0.57
    ment
    0.57
    Activations Density 0.000%

    No Known Activations