© 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. 1028
    Prev
    Next
    INDEX
    Explanations

    MMORPG

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

    This neuron fires on isolated uppercase letters used as initials or abbreviations (e.g. single capital letters followed by periods).

    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
    しか
    0.46
    用意
    0.45
    変数
    0.44
    很
    0.44
     نمی
    0.44
    लेक्ट
    0.44
     nonempty
    0.43
    ص
    0.42
    踏
    0.42
    Especific
    0.42
    POSITIVE LOGITS
     Pension
    0.60
     Pensions
    0.58
     T
    0.56
     Hockey
    0.55
     billionaire
    0.52
     M
    0.50
     District
    0.50
     Commissioner
    0.50
     National
    0.49
     Regional
    0.49
    Activations Density 0.001%

    No Known Activations