© 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. 108968
    Prev
    Next
    INDEX
    Explanations

    two-letter codes

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

    This neuron fires on standalone uppercase letter sequences—often two-letter acronyms or initialisms—especially when they include digits.

    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
    <bos>
    -0.90
    Zubereitung
    -0.90
    kå
    -0.89
    でしょうね
    -0.88
    ？』
    -0.84
     stazione
    -0.83
    locke
    -0.83
     PLEASE
    -0.83
    ázky
    -0.82
    Elaboración
    -0.82
    POSITIVE LOGITS
    Nik
    1.02
     dansk
    0.94
    ug
    0.91
    Flor
    0.89
    蹿
    0.88
    ícios
    0.87
     Cigar
    0.86
    orf
    0.85
    opher
    0.85
    Recon
    0.84
    Activations Density 0.054%

    No Known Activations