© 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. 34-GEMMASCOPE-RES-131K
    7. 48209
    Prev
    Next
    INDEX
    Explanations

    tells you

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

    The neuron is picking out very high‐frequency “common” tokens—especially stop words, punctuation marks, and other ubiquitous short pieces like “the,” “help,” “give,” “.”, hyphens, etc.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     homeschool
    -0.80
    นะ
    -0.76
    Multiplication
    -0.75
     wife
    -0.75
    John
    -0.70
     Multiplication
    -0.70
    Rick
    -0.70
     njego
    -0.69
    David
    -0.69
     families
    -0.68
    POSITIVE LOGITS
    igel
    0.76
     Bowser
    0.76
    KURZBESCHREIBUNG
    0.76
    Տես
    0.76
     immortality
    0.75
     Probl
    0.75
    jint
    0.75
    atla
    0.75
     werk
    0.74
    得罪
    0.74
    Activations Density 0.013%

    No Known Activations