© 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. 31-GEMMASCOPE-2-RES-16K
    7. 6628
    Prev
    Next
    INDEX
    Explanations

    The neuron fires strongly on numeric tokens and number‐like sequences (e.g. digits, percentages, or numeric codes).

    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_31_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
     .'
    0.71
     তাহার
    0.63
     ,'
    0.60
     .-
    0.59
     είχε
    0.58
     .$
    0.56
     चळ
    0.56
     Untersuchung
    0.55
     น
    0.55
     τό
    0.55
    POSITIVE LOGITS
    ormal
    0.59
    Thankfully
    0.56
    dioxide
    0.56
    ்த்து
    0.55
    rekking
    0.55
    γγελμα
    0.53
    рито
    0.52
    他にも
    0.52
    prost
    0.51
    ギュラー
    0.51
    Activations Density 0.002%

    No Known Activations