© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 38500
    Prev
    Next
    INDEX
    Explanations

    defence ministry

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

    The neuron strongly activates on named‐entity and numeric tokens—i.e. organization and location names, people’s names, dates, and standalone numbers.

    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
    ემ
    -0.84
     stopped
    -0.76
    edipus
    -0.76
     called
    -0.75
     neither
    -0.75
    を選択します
    -0.75
     lower
    -0.73
     Nema
    -0.73
     singoli
    -0.72
     внутріш
    -0.71
    POSITIVE LOGITS
    abler
    0.77
     batalla
    0.76
    drawImage
    0.75
    Semaphore
    0.75
     Oscars
    0.75
     CLI
    0.75
     reais
    0.75
     Induction
    0.74
     BCG
    0.73
    vspace
    0.72
    Activations Density 0.013%

    No Known Activations