© 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. 660
    Prev
    Next
    INDEX
    Explanations

    best given films

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

    The neuron primarily activates on document metadata and headings—lines of all-caps or title-style text (e.g. section names, bibliographic titles, copyright lines, numbered headings).

    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
    blr
    -0.81
    imperial
    -0.79
    %*
    -0.78
     šo
    -0.77
    joo
    -0.77
    انظر
    -0.75
    fmf
    -0.75
    
    -0.74
    ioi
    -0.74
     savu
    -0.74
    POSITIVE LOGITS
    uities
    0.80
     things
    0.78
     ttk
    0.77
    izeit
    0.75
    購入
    0.75
    yafet
    0.74
    去世
    0.74
    pdf
    0.74
     APIs
    0.73
    ged
    0.73
    Activations Density 0.009%

    No Known Activations