© 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. 73187
    Prev
    Next
    INDEX
    Explanations

    Mustafa, Changes, Solid

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

    The neuron flags Latin‐script words of Arabic origin (i.e. transliterated Arabic terms like Esteghlal, Mujtahid, Istiqlal, Mutazila, Intifada, 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
    '
    -3.05
     When
    -2.28
     The
    -2.22
    </b>
    -2.13
    </i>
    -2.11
    )
    -1.98
     cambian
    -1.96
     if
    -1.96
    でも
    -1.95
    me
    -1.91
    POSITIVE LOGITS
    珦
    3.13
    囖
    2.89
    脔
    2.86
    
    2.69
    ครื่อง
    2.55
    蛏
    2.55
     aching
    2.50
    橚
    2.48
    蒾
    2.45
    踅
    2.38
    Activations Density 0.002%

    No Known Activations