© 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. 34-GEMMASCOPE-RES-131K
    7. 75900
    Prev
    Next
    INDEX
    Explanations

    erotic, erosive, sirolimus, oropharyngeal

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

    words starting with "ero," "siro," "oro," "iro," or similar phonetic patterns.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    This neuron fires strongly on any occurrence of the character sequence “ro” (often spanning across token boundaries) in a word.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    應該
    -0.75
     осуществ
    -0.75
     correct
    -0.69
     r
    -0.69
    publ
    -0.69
    ！！！”
    -0.68
     здійс
    -0.67
    issants
    -0.66
     quick
    -0.66
    gura
    -0.66
    POSITIVE LOGITS
    almi
    0.85
    υχ
    0.81
    sige
    0.80
    sphinx
    0.79
    traditional
    0.79
     autorización
    0.78
    eror
    0.77
    escent
    0.76
    哮
    0.75
     saks
    0.75
    Activations Density 0.087%

    No Known Activations