© 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
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 10114
    Prev
    Next
    INDEX
    Explanations

    mile east

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

    This neuron fires strongly at tokens that mark major structural boundaries (e.g. the start of a new section or list after a blank line).

    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
    स
    0.89
    fica
    0.88
    ó
    0.82
    蠻
    0.82
     infractions
    0.79
    ع
    0.79
    عين
    0.78
    acoes
    0.77
    ਮ
    0.77
    攻擊
    0.76
    POSITIVE LOGITS
     Maier
    0.79
     dagger
    0.73
     appId
    0.73
     dandelion
    0.72
     מיט
    0.72
     disk
    0.71
     swelled
    0.71
     swan
    0.71
    Analyzer
    0.70
     paraf
    0.70
    Activations Density 0.000%

    No Known Activations