© 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. 10-GEMMASCOPE-RES-131K
    4. 99630
    Prev
    Next
    INDEX
    Explanations

    TNF Block gene variants

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

    postal codes, PO box numbers, and zip codes.

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

    The neuron primarily activates on numeric tokens or number sequences (digits, decimals, 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_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
     bluza
    -1.79
    Its
    -1.70
    
    -1.68
     koszulka
    -1.60
     asegurarse
    -1.60
     endTime
    -1.60
    などなど
    -1.59
    许久
    -1.59
    ㍉
    -1.59
    ⃠
    -1.58
    POSITIVE LOGITS
     of
    1.78
    ↵
    1.68
     for
    1.66
     from
    1.64
     altre
    1.63
     Each
    1.60
     boxes
    1.56
     to
    1.51
     Then
    1.45
    Acacia
    1.43
    Activations Density 0.004%

    No Known Activations