© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 100869
    Prev
    Next
    INDEX
    Explanations

    problem, past, Parliament, Allah, everything, world

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

    The neuron fires on multiword, topic‐defining phrases—especially capitalized, domain‐specific nouns (as in headings, titles, or technical subject labels).

    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
     in
    -1.77
     like
    -1.70
     through
    -1.70
     new
    -1.52
     from
    -1.44
     with
    -1.32
     menyiapkan
    -1.27
    しちゃ
    -1.26
     along
    -1.25
     traditionellen
    -1.20
    POSITIVE LOGITS
     who
    1.50
    那些
    1.38
     pastillas
    1.36
    )}=
    1.34
    
    1.30
     должны
    1.26
     cuanto
    1.23
    anda
    1.23
     cuadrados
    1.23
     roj
    1.23
    Activations Density 0.293%

    No Known Activations