© 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. 10-GEMMASCOPE-RES-131K
    7. 105246
    Prev
    Next
    INDEX
    Explanations

    friendly followed by a noun

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

    The neuron activates on hyphenated “-friendly” descriptors (e.g. family-friendly, kid-friendly, newbie-friendly, 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_10/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
     you
    -1.80
     there
    -1.54
    нии
    -1.52
     pacar
    -1.47
     some
    -1.47
     kedu
    -1.45
     gelar
    -1.45
     =
    -1.44
     continúas
    -1.38
     televisiva
    -1.36
    POSITIVE LOGITS
    又不
    1.85
    0
    1.75
    楹
    1.65
    却不
    1.64
    ₶
    1.62
     suoi
    1.61
    Its
    1.59
     saine
    1.49
    瓮
    1.47
     ۱۴
    1.46
    Activations Density 0.007%

    No Known Activations