© 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. 34-GEMMASCOPE-RES-131K
    7. 82379
    Prev
    Next
    INDEX
    Explanations

    in advance

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

    The neuron fires on key content words—especially nouns and verbs that carry topic-specific meaning (like “prepared,” “processing,” “screening,” “data,” “research,” etc.)—rather than on common function words.

    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
    naments
    -0.90
    луб
    -0.83
    ical
    -0.82
    Past
    -0.78
    bt
    -0.78
     finalement
    -0.77
    ko
    -0.77
     definitiva
    -0.76
    ICAL
    -0.76
    sivamente
    -0.75
    POSITIVE LOGITS
     ahead
    5.59
     beforehand
    4.88
    ahead
    4.13
     Ahead
    3.97
     заранее
    3.92
    Ahead
    3.88
    提前
    3.42
     AHEAD
    3.36
     prior
    3.13
     vooraf
    2.98
    Activations Density 0.128%

    No Known Activations