© 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. 15050
    Prev
    Next
    INDEX
    Explanations

    erotica, gender, stereotypes, authentication

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

    The neuron fires most strongly on words immediately preceding decimal‐formatted numbers, i.e. it marks contexts where a numeric measurement (a floating-point value) follows.

    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
    chaft
    0.88
    y
    0.82
    म्मू
    0.79
     Cohen
    0.79
     Pandya
    0.76
     وعلى
    0.76
    yó
    0.75
    ine
    0.75
     Shroff
    0.75
     Recher
    0.74
    POSITIVE LOGITS
     critiques
    0.86
    тити
    0.81
    น
    0.81
     verbs
    0.80
     proof
    0.80
    উনি
    0.79
    ট
    0.79
    ထ
    0.77
    т
    0.76
    न
    0.75
    Activations Density 0.000%

    No Known Activations