© 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. 104941
    Prev
    Next
    INDEX
    Explanations

    company or specific terms

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

    The neuron fires on multiword noun or descriptive collocations (e.g. “thought processes,” “floor to ceiling”).

    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
     to
    -1.62
     compacta
    -1.59
    鹀
    -1.56
    !!!!!!!!!!!!!!!!
    -1.55
     would
    -1.55
     prega
    -1.55
     This
    -1.53
     characterized
    -1.51
    しかも
    -1.50
     prendra
    -1.49
    POSITIVE LOGITS
    ists
    1.77
    --
    1.65
    ,
    1.62
    5
    1.55
    央视
    1.51
    ใกล้
    1.50
     användas
    1.47
    ㉕
    1.47
     jiné
    1.46
    かなり
    1.45
    Activations Density 0.027%

    No Known Activations