© 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. 22-GEMMASCOPE-RES-131K
    7. 14203
    Prev
    Next
    INDEX
    Explanations

    other and also

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

    The neuron spikes strongly on tokens that mark sentence boundaries—namely the initial token of a sentence and the final punctuation at its end.

    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
    だったので
    -1.16
     которых
    -1.09
     обычно
    -1.08
    isDeleted
    -1.06
     сначала
    -1.05
     trámite
    -1.05
    companyName
    -1.04
     będzie
    -1.02
     hinten
    -1.00
     тільки
    -0.99
    POSITIVE LOGITS
     other
    5.66
     otros
    3.61
     diğer
    3.58
     других
    3.52
     anderen
    3.44
     autres
    3.44
     others
    3.42
    其他
    3.39
     also
    3.38
    autres
    3.38
    Activations Density 0.038%

    No Known Activations