© 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. 12948
    Prev
    Next
    INDEX
    Explanations

    This neuron strongly activates on the final substantive word of each text segment (i.e. the last content-bearing token before a break).

    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
     decades
    0.58
    쩡
    0.52
     while
    0.50
    Bottom
    0.49
     SizedBox
    0.49
     Tutak
    0.48
     Bengaluru
    0.48
     cities
    0.48
    십
    0.47
    FromArgb
    0.46
    POSITIVE LOGITS
    いろんな
    0.60
    いろいろ
    0.59
    ش
    0.56
    exposition
    0.55
     ей
    0.55
     покер
    0.53
     μεγάλη
    0.52
    幣
    0.52
     intensa
    0.51
    פ
    0.51
    Activations Density 0.000%

    No Known Activations