© 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. 22-GEMMASCOPE-RES-131K
    7. 25919
    Prev
    Next
    INDEX
    Explanations

    introduction to

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

    The neuron fires on very common, high-frequency function words and punctuation that characterize generic, neutral prose.

    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
     my
    -1.52
     ยาว
    -1.23
     these
    -1.23
    ଓ
    -1.21
    картинка
    -1.20
     cukru
    -1.19
     сахара
    -1.18
     marah
    -1.17
     sendok
    -1.16
     дуже
    -1.16
    POSITIVE LOGITS
    get
    1.23
    ube
    1.17
    un
    1.09
    ˆ
    1.09
    If
    1.07
    are
    1.06
    还不错
    1.06
    ></
    1.05
    看来
    1.05
    adren
    1.04
    Activations Density 0.002%

    No Known Activations