© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 9274
    Prev
    Next
    INDEX
    Explanations

    via text message

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

    The neuron is specialized to detect decimal numbers (numerical tokens containing a period).

    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
    Ông
    0.63
    壑
    0.59
     방정
    0.56
    it
    0.55
    पयोग
    0.55
    ﺤ
    0.55
    사를
    0.54
    어진
    0.53
     GoName
    0.53
     入っ
    0.52
    POSITIVE LOGITS
     Р
    0.68
     llevó
    0.67
    і
    0.65
     llevaron
    0.61
    К
    0.61
    ной
    0.59
     ר
    0.58
     doppia
    0.57
    ר
    0.57
     ين
    0.57
    Activations Density 0.001%

    No Known Activations