© 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. 6437
    Prev
    Next
    INDEX
    Explanations

    shut down

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

    This neuron consistently activates on the final salient content word (noun or verb) at the end of a segment or sentence.

    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
    ون
    0.86
     extremos
    0.79
    𝕥
    0.79
    éal
    0.78
    𝓎
    0.77
     deutlich
    0.76
     chauss
    0.76
    जीत
    0.75
    ቀ
    0.75
     selanjutnya
    0.75
    POSITIVE LOGITS
     Гра
    0.72
     الانسان
    0.71
     Neg
    0.70
     نگر
    0.70
     Posts
    0.69
     AppCompat
    0.66
     humanist
    0.66
     Months
    0.66
    ள்ளதாக
    0.65
     الإنسان
    0.65
    Activations Density 0.000%

    No Known Activations