© 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. 10-GEMMASCOPE-RES-131K
    7. 57366
    Prev
    Next
    INDEX
    Explanations

    original item descriptions

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

    The neuron fires strongly on the key technical subject or “head” noun in a question or description (e.g. pointer, error, function, circle, date, database, etc.), effectively spotlighting the main topic word.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     aussit
    -2.06
     rése
    -2.00
     zwi
    -1.99
    傒
    -1.91
     appeler
    -1.90
     Autó
    -1.85
     uyt
    -1.84
     Atención
    -1.84
     miniaturka
    -1.83
     procha
    -1.78
    POSITIVE LOGITS
     is
    2.20
     or
    2.19
     on
    1.98
     two
    1.91
     through
    1.87
     p
    1.84
     because
    1.82
     but
    1.80
     had
    1.77
     has
    1.76
    Activations Density 0.184%

    No Known Activations