© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 64263
    Prev
    Next
    INDEX
    Explanations

    font weights and styles

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

    The neuron strongly activates on isolated numeric constants or version‐style numbers (e.g. years, counts, version tags) appearing as standalone 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_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
     UIColor
    -0.82
    坎
    -0.81
    oane
    -0.81
     NEED
    -0.78
     disappear
    -0.78
    scoped
    -0.77
     disappears
    -0.76
    Like
    -0.75
    幹線
    -0.75
    Much
    -0.75
    POSITIVE LOGITS
    bold
    1.12
     FontWeight
    0.95
     bold
    0.92
    Bold
    0.89
    BOLD
    0.88
    w
    0.88
     Bold
    0.88
     W
    0.86
     normal
    0.82
    Thin
    0.81
    Activations Density 0.000%

    No Known Activations