© 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. 34-GEMMASCOPE-RES-131K
    7. 1082
    Prev
    Next
    INDEX
    Explanations

    screen and related prefixes

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

    The neuron is picking out emphatic, attention‐grabbing tokens—especially exclamation marks, numeric values, and other strong punctuation or uppercase “shouty” markers.

    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
    rough
    -0.85
     historischen
    -0.84
     podría
    -0.82
     historische
    -0.82
     junio
    -0.79
    izz
    -0.78
    century
    -0.78
    был
    -0.77
    oregon
    -0.77
    five
    -0.76
    POSITIVE LOGITS
     screened
    0.85
    コミュニティ
    0.82
     Tiles
    0.79
     Screen
    0.78
    屏幕
    0.77
     Upgrades
    0.77
    SCREEN
    0.76
     geme
    0.75
    Screen
    0.74
     Stru
    0.74
    Activations Density 0.010%

    No Known Activations