© 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. 39452
    Prev
    Next
    INDEX
    Explanations

    X-Men #95

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

    The neuron fires on document header/title metadata tokens—things like all-caps words, hyphens, single-letter capital abbreviations, and numbers that typically appear in headings.

    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
    charpe
    -0.91
    原始内容存档于
    -0.86
     Einheit
    -0.86
    dreams
    -0.85
    '])->
    -0.84
    cribed
    -0.82
     Dreams
    -0.81
     initialise
    -0.81
    ✗
    -0.81
     wszel
    -0.79
    POSITIVE LOGITS
     麺
    0.98
     applicazione
    0.90
    appré
    0.84
    шна
    0.83
     reflectance
    0.82
    белый
    0.81
     produite
    0.81
    貶
    0.81
     宽
    0.80
     kuu
    0.79
    Activations Density 0.004%

    No Known Activations