© 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. 66037
    Prev
    Next
    INDEX
    Explanations

    names with following particles

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

    This neuron responds to occurrences of the substring “hao,” as in the name “Hao.”

    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
     Aufenthalt
    -2.69
     beinahe
    -2.69
     erwähnten
    -2.63
     That
    -2.59
     höheren
    -2.58
    entarios
    -2.58
    sibirsk
    -2.55
     beobachtet
    -2.53
     bemerkt
    -2.53
    selben
    -2.52
    POSITIVE LOGITS
    a
    3.02
    x
    2.77
     είναι
    2.70
     sont
    2.61
    IN
    2.47
    就开始
    2.39
    k
    2.38
     constitue
    2.36
    独特的
    2.33
     própri
    2.28
    Activations Density 0.001%

    No Known Activations