© 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. 2990
    Prev
    Next
    INDEX
    Explanations

    particularly

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

    The neuron triggers on unusually large numeric tokens (high-magnitude numbers with decimal values).

    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
    ^{-}
    0.80
    ranks
    0.75
    ✬
    0.75
     krát
    0.73
     राज्यों
    0.72
     ['.
    0.71
    ্ভব
    0.70
    movies
    0.70
     spôsob
    0.69
     notoriously
    0.68
    POSITIVE LOGITS
    ach
    0.91
    ↵
    0.88
    ске
    0.76
     Federation
    0.76
    荣誉
    0.75
    ग्
    0.72
    ar
    0.68
    rie
    0.68
    apping
    0.68
    itz
    0.68
    Activations Density 0.000%

    No Known Activations