© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 28568
    Prev
    Next
    INDEX
    Explanations

    Diana Gabaldon's quotes

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

    The neuron strongly activates on word‐fragments that include or end in punctuation affixes—specifically tokens that carry trailing hyphens or apostrophes (as in “roller-” or “’s”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     teammates
    -0.88
     diferente
    -0.84
     userDao
    -0.82
     воспита
    -0.82
    },"
    -0.81
     раскры
    -0.78
    ######
    -0.78
     known
    -0.77
     разные
    -0.77
    different
    -0.77
    POSITIVE LOGITS
     laajenta
    0.99
    到来
    0.93
    そして
    0.91
    ższy
    0.91
    Postup
    0.91
    ిక
    0.85
    glGet
    0.84
    AntiForgeryToken
    0.84
     aqueles
    0.84
    inerario
    0.84
    Activations Density 0.009%

    No Known Activations