© 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. 111280
    Prev
    Next
    INDEX
    Explanations

    names like Royce Classic, Derrick Watson

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

    The neuron is triggered by proper names and technical acronyms (i.e. named‐entity 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
    2
    -3.03
    8
    -2.80
    3
    -2.78
    是非常
    -2.70
     setelah
    -2.69
    $.
    -2.67
     Besuch
    -2.64
     Schwierigkeiten
    -2.64
     we
    -2.59
     wonders
    -2.59
    POSITIVE LOGITS
    ᠁
    2.95
    
    2.70
    ]
    2.58
    ”(
    2.55
    珖
    2.45
    撘
    2.42
    洐
    2.41
    ”?
    2.39
    錒
    2.39
    ’
    2.38
    Activations Density 0.016%

    No Known Activations