© 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. 3244
    Prev
    Next
    INDEX
    Explanations

    The neuron is detecting numeric tokens with decimal points—specifically floating‐point numbers (e.g. years or other four-digit values with decimals).

    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
    л
    0.82
     эк
    0.80
    emper
    0.79
    с
    0.75
     estimado
    0.74
     detalles
    0.74
     мело
    0.73
    ERTY
    0.73
     Mechanic
    0.73
     passphrase
    0.72
    POSITIVE LOGITS
    كة
    0.80
    Katie
    0.73
    หญิง
    0.70
    aithe
    0.69
    ตั้งแต่
    0.69
    หลากหลาย
    0.69
    闘
    0.67
    ऑस्ट्रेलिया
    0.66
    लेंट
    0.65
     persu
    0.64
    Activations Density 0.000%

    No Known Activations