© 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. 105292
    Prev
    Next
    INDEX
    Explanations

    arising from

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

    The neuron fires on occurrences of the verb “arise” (in forms like “arising” or “arises”).

    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
    ↵
    -2.67
    \
    -2.55
    .
    -2.41
    Y
    -2.25
    g
    -2.25
    E
    -2.17
    0
    -2.16
    蜴
    -2.16
    F
    -2.11
     pero
    -2.09
    POSITIVE LOGITS
    
    2.53
     fragte
    2.41
    amerikanische
    2.33
     ")";
    2.28
     подключения
    2.23
    𝓹
    2.16
    Spesifikasi
    2.16
    ?”
    2.14
    LLABLE
    2.13
     别墅
    2.11
    Activations Density 0.007%

    No Known Activations