© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 73532
    Prev
    Next
    INDEX
    Explanations

    name beginnings

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

    This neuron fires on capitalized tokens that are parts of named entities—particularly proper nouns in place and personal names.

    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
    むしろ
    -1.10
    Pró
    -0.97
    četně
    -0.95
     eradicate
    -0.91
    最強
    -0.90
     kontinuier
    -0.89
    初期
    -0.88
     القيام
    -0.88
     gänzlich
    -0.88
     Grüßen
    -0.87
    POSITIVE LOGITS
    ølge
    1.27
    alties
    1.23
    ¹)
    1.06
    voer
    1.05
    romeda
    1.02
    –
    1.02
    lably
    0.98
    spective
    0.98
    ês
    0.98
    ScopeManager
    0.97
    Activations Density 0.025%

    No Known Activations