© 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. 5323
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on numeric tokens (digits, decimal numbers, measurements, phone numbers, etc.).

    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
    ک
    0.83
    看到
    0.80
    夏天
    0.79
    போது
    0.79
    సూరు
    0.79
    貅
    0.79
    應該
    0.78
    দেখে
    0.78
     métallique
    0.77
    国内
    0.76
    POSITIVE LOGITS
     B
    0.82
    ans
    0.82
     Tycho
    0.79
    ing
    0.79
    T
    0.78
    j
    0.75
     On
    0.75
     P
    0.74
    ta
    0.74
    ut
    0.74
    Activations Density 0.000%

    No Known Activations