© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 68399
    Prev
    Next
    INDEX
    Explanations

    blank

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

    This neuron fires strongly on “Blank” tokens—that is, placeholders marking blank or empty content (e.g. empty lines or large spans of whitespace).

    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
     แต่
    -3.61
    难以
    -2.75
    ről
    -2.73
     欣赏
    -2.66
     stirs
    -2.63
    ではある
    -2.61
     monstrous
    -2.58
     wist
    -2.58
     bezpeč
    -2.56
    翃
    -2.55
    POSITIVE LOGITS
    i
    3.48
    which
    3.20
    This
    3.14
    4
    3.06
    ]
    2.95
    0
    2.91
     или
    2.91
    इटम
    2.89
    Most
    2.88
    All
    2.84
    Activations Density 0.007%

    No Known Activations