© 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. 3725
    Prev
    Next
    INDEX
    Explanations

    path, split, getResource, cache

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

    The neuron is triggered by capitalized identifier fragments—i.e. the CamelCase or PascalCase subwords (class/method/annotation/config names) in code.

    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
    瑧
    -1.57
    $
    
    -1.48
     különböz
    -1.45
    $}
    -1.42
    Accademia
    -1.41
    fani
    -1.41
    始める
    -1.41
     atencion
    -1.39
    德里
    -1.34
    ]**
    -1.34
    POSITIVE LOGITS
     are
    1.61
    <bos>
    1.47
     justa
    1.29
     ‘
    1.27
     oberen
    1.26
     bedeutend
    1.23
     ´
    1.22
     volle
    1.21
    umps
    1.20
     sech
    1.17
    Activations Density 0.010%

    No Known Activations