© 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. 34-GEMMASCOPE-RES-131K
    7. 49208
    Prev
    Next
    INDEX
    Explanations

    specific token followed by common continuations

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

    The neuron spikes most strongly on the very first meaningful content words (and numbers) at the start of each document or segment.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     является
    -1.06
     также
    -0.88
    Additionally
    -0.81
     Additionally
    -0.79
     أيضا
    -0.78
     également
    -0.76
     Fuck
    -0.73
    ческой
    -0.72
    Since
    -0.72
    🏢
    -0.71
    POSITIVE LOGITS
     võib
    0.85
    minimal
    0.84
     wouldn
    0.82
    ⑹
    0.80
    大丈夫です
    0.80
     isn
    0.79
    imientos
    0.78
    tyw
    0.78
    chtige
    0.77
    Bedankt
    0.77
    Activations Density 0.002%

    No Known Activations