© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 80030
    Prev
    Next
    INDEX
    Explanations

    translation strings

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

    The neuron fires on numeric literals, especially numbers used for position‐metadata fields (like “line”, “column”, and “end”) in code or JSON.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     low
    -1.20
     receive
    -1.06
     work
    -0.97
    ہ
    -0.96
     if
    -0.95
     earlier
    -0.95
     good
    -0.94
    IFORN
    -0.92
     globally
    -0.91
    ançon
    -0.90
    POSITIVE LOGITS
    atize
    1.00
     евре
    0.95
     conster
    0.92
    璆
    0.91
     recens
    0.90
     باستخدام
    0.89
    甾
    0.89
     間取り
    0.89
     levé
    0.89
    と思っている
    0.88
    Activations Density 0.001%

    No Known Activations