© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 43684
    Prev
    Next
    INDEX
    Explanations

    proven effectiveness

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

    The neuron activates on claims of verified performance, especially the word “proven” (often in “proven track record”).

    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
     fotografie
    -2.23
    푘
    -2.08
    爌
    -2.08
     tantangan
    -2.05
    驺
    -1.97
    茕
    -1.95
     bekannten
    -1.93
    -------
    -1.89
    ࿊
    -1.89
    
    -1.88
    POSITIVE LOGITS
    ’
    3.14
     –
    2.98
    ’,
    2.73
    );
    2.59
    It
    2.56
    ,”
    2.50
    ↵
    2.31
    它
    2.28
    .’
    2.28
    一旁
    2.27
    Activations Density 0.004%

    No Known Activations