© 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. 22-GEMMASCOPE-RES-131K
    4. 75186
    Prev
    Next
    INDEX
    Explanations

    materials and methods

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

    section headers or structural elements in scientific papers, particularly methodology-related headings.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron is picking out numeric tokens (digits and numbers) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     that
    -1.48
     to
    -1.41
     if
    -1.34
     see
    -1.09
    </tfoot>
    -1.05
     apabila
    -1.05
     always
    -1.00
     when
    -0.98
     растения
    -0.97
     If
    -0.96
    POSITIVE LOGITS
    gebnis
    1.10
     البحث
    1.08
    graphiques
    1.05
     firestore
    1.05
    RESULTS
    1.05
     이용
    0.98
    FONTE
    0.97
    解码
    0.97
     perfecte
    0.97
    マウス
    0.97
    Activations Density 0.180%

    No Known Activations