© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 98588
    Prev
    Next
    INDEX
    Explanations

    underscore followed by word

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

    The neuron’s primary role is to detect underscore characters ( “_” ) in the text.

    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
     she
    -1.22
     others
    -0.98
     shows
    -0.90
    きり
    -0.89
     kopling
    -0.88
    EIO
    -0.86
    のように
    -0.85
    があります
    -0.85
     怎么
    -0.84
    ねぇ
    -0.83
    POSITIVE LOGITS
     _
    1.09
    ,_
    0.99
    ༀ
    0.98
    ときは
    0.97
     _)
    0.94
    }}$,
    0.93
    =_
    0.93
    ::_
    0.92
    本日の
    0.92
    ._
    0.91
    Activations Density 0.020%

    No Known Activations