© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 127781
    Prev
    Next
    INDEX
    Explanations

    numbers with decimal points

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

    The neuron is highly sensitive to the digit “2” appearing as part of numerical values.

    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
     to
    -2.06
     by
    -2.03
    IANGLES
    -2.00
     it
    -1.98
    沕
    -1.95
    他对
    -1.94
     believes
    -1.92
     trivi
    -1.85
    銪
    -1.85
    趿
    -1.84
    POSITIVE LOGITS
    2
    4.00
    u
    2.70
    I
    2.53
    nd
    2.34
    v
    2.33
    3
    2.27
    E
    2.22
    
    2.22
    """
    2.20
    存知
    2.11
    Activations Density 0.027%

    No Known Activations