© 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. 72762
    Prev
    Next
    INDEX
    Explanations

    special characters and words

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

    The neuron fires on punctuation and symbol‐rich tokens typical of code/math or markup syntax (e.g. “$”, “(”, “)”, “<”, “>”, “=”, identifiers like f, x, hyphenated names, etc.).

    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
    gi
    -0.89
    에서
    -0.81
    按照
    -0.79
     ferner
    -0.79
    Z
    -0.78
    li
    -0.77
    nu
    -0.75
    ta
    -0.74
    జ
    -0.73
     geplant
    -0.72
    POSITIVE LOGITS
    Ưu
    0.94
     miliki
    0.85
    nước
    0.85
    Ngoài
    0.83
    Cuándo
    0.81
    Тан
    0.80
    ByDefault
    0.80
    ahnya
    0.80
     černá
    0.79
    ieval
    0.79
    Activations Density 0.164%

    No Known Activations