© 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. 84099
    Prev
    Next
    INDEX
    Explanations

    definitely followed by various words

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

    The neuron is detecting emphatic adverbs that convey strong certainty or emphasis (e.g. “definitely,” “always,” “absolutely”).

    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
     één
    -1.60
    走吧
    -1.59
     کہ
    -1.47
    垕
    -1.47
    ére
    -1.47
     съдържа
    -1.46
    documento
    -1.45
    푉
    -1.45
    Descripció
    -1.44
    
    -1.44
    POSITIVE LOGITS
    ’
    1.83
    ходить
    1.63
    '
    1.62
     whose
    1.61
    which
    1.59
     would
    1.52
     which
    1.52
    e
    1.49
     allow
    1.48
    p
    1.48
    Activations Density 0.007%

    No Known Activations