© 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. 10-GEMMASCOPE-RES-131K
    7. 66275
    Prev
    Next
    INDEX
    Explanations

    pers followed by words

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

    The neuron consistently activates on the character sequence “pers” (i.e. it detects the substring “pers” within tokens).

    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
     
    -3.20
    ↵
    -3.00
     was
    -2.81
     be
    -2.80
     We
    -2.48
     Are
    -2.42
     he
    -2.34
     ”
    -2.31
     l
    -2.27
    もあった
    -2.25
    POSITIVE LOGITS
    栊
    2.95
    猡
    2.75
     内衣
    2.63
    鯔
    2.59
     贺
    2.59
    ﾙ
    2.58
    
    2.56
    뿅
    2.55
     thc
    2.53
     套装
    2.53
    Activations Density 0.003%

    No Known Activations