© 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. 12196
    Prev
    Next
    INDEX
    Explanations

    good intentions and character

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

    The neuron fires on words that express positive intentions or goodwill (e.g. “well-meaning,” “helpful,” “good”)—i.e. benevolent motivations.

    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
    与此同时
    -1.59
    嗄
    -1.44
     OGSÅ
    -1.43
    除此
    -1.37
    我和
    -1.36
     determining
    -1.35
    ｋ
    -1.34
    继续阅读
    -1.34
    最后由
    -1.34
    ಞ
    -1.32
    POSITIVE LOGITS
    ↵↵
    2.13
    "
    1.78
    ',
    1.55
    ,”
    1.50
    0
    1.46
    ",
    1.41
    這個
    1.40
     —
    1.38
    ,.
    1.38
    もいい
    1.34
    Activations Density 0.013%

    No Known Activations