© 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. 73310
    Prev
    Next
    INDEX
    Explanations

    Mat followed by lab, ploit, hemat, tson

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

    The neuron activates on tokens containing the character sequence “Mat,” effectively detecting that specific three‐letter substring in words.

    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
    ÇÕES
    -0.88
    小山
    -0.80
     incredible
    -0.78
     besitzen
    -0.77
    ponential
    -0.75
    clusion
    -0.75
     incapacity
    -0.74
    恣
    -0.73
     Michael
    -0.72
    矯正
    -0.71
    POSITIVE LOGITS
     Matos
    0.96
    mat
    0.95
    Mat
    0.93
     mater
    0.91
     mat
    0.91
    mati
    0.87
    mata
    0.87
    わざわざ
    0.87
    だいぶ
    0.85
    ​​​
    0.85
    Activations Density 0.014%

    No Known Activations