© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 36609
    Prev
    Next
    INDEX
    Explanations

    exploration expeditions and surveys

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

    The neuron spotlights structured metadata entries—field names (like “Survey,” “Project,” “Duration”) and their associated numerical values (dates, years, measurements)—as found in infobox-style or technical summary sections.

    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
     as
    -1.44
     like
    -1.41
     ဝ
    -1.38
     any
    -1.37
     your
    -1.33
     inspirasi
    -1.28
     sosok
    -1.27
    醯
    -1.27
    饷
    -1.24
    帐户
    -1.19
    POSITIVE LOGITS
    ֵּ
    1.42
     planned
    1.41
     conducted
    1.35
    ֶּ
    1.33
     to
    1.32
     undertaken
    1.30
    落在
    1.26
     untersucht
    1.23
    спеди
    1.19
     ベクトル
    1.18
    Activations Density 0.039%

    No Known Activations