© 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. 34-GEMMASCOPE-RES-131K
    7. 3404
    Prev
    Next
    INDEX
    Explanations

    comments and views

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

    The neuron is primarily detecting standalone numeric metadata (e.g. comment counts, view counts) 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
     Besøkt
    -0.88
     this
    -0.85
    ницей
    -0.82
     At
    -0.79
     it
    -0.79
     quite
    -0.78
    because
    -0.78
     just
    -0.78
     Data
    -0.77
    Hello
    -0.76
    POSITIVE LOGITS
     reply
    1.07
     replies
    0.94
    返信
    0.94
    replies
    0.92
     replied
    0.92
     femur
    0.84
     humer
    0.84
    reply
    0.83
    edited
    0.83
     バッジ
    0.83
    Activations Density 0.001%

    No Known Activations