© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 4827
    Prev
    Next
    INDEX
    Explanations

    given/each/following item

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

    The neuron primarily activates on numeric tokens, especially list‐item or question numbering.

    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
    byshire
    0.59
     又
    0.57
    一支
    0.56
     Nên
    0.56
     佐
    0.55
    raham
    0.55
     Cham
    0.55
     dụng
    0.54
    Cham
    0.54
     秋
    0.54
    POSITIVE LOGITS
    などは
    0.65
    ങ്ങളിലും
    0.63
    ത്തിലും
    0.62
    实在是
    0.59
    いずれ
    0.56
    স্থলে
    0.56
     agricult
    0.56
    .')
    0.56
     philanthrop
    0.56
    .'),
    0.55
    Activations Density 0.000%

    No Known Activations