© 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. 3579
    Prev
    Next
    INDEX
    Explanations

    Explanation of neuron 4 behavior: the main thing this neuron does is find floating‐point numeric expressions.

    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
     the
    0.40
    й
    0.35
    一覧
    0.33
     this
    0.32
    }=\
    0.32
     ardından
    0.31
     istri
    0.31
    }}=\
    0.31
    }}=-\
    0.30
     الجديد
    0.30
    POSITIVE LOGITS
    hundred
    0.50
     тысяч
    0.50
     thousand
    0.46
     kilow
    0.46
     trillion
    0.45
     kilowatt
    0.44
     दर्जन
    0.44
     كيلو
    0.43
     hundred
    0.43
     kilobytes
    0.42
    Activations Density 0.129%

    No Known Activations