© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 7058
    Prev
    Next
    INDEX
    Explanations

    projected, projection, projected, projecting

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

    The neuron selectively fires on LaTeX/math‐mode operator and delimiter tokens (e.g. “\mapsto,” “\int,” “||,” “\phi,” etc.).

    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
     verhalen
    -0.88
    anglès
    -0.85
     betekenis
    -0.84
     verschillen
    -0.84
     lehe
    -0.83
     redenen
    -0.81
     experiência
    -0.81
    Sentiment
    -0.80
     aanzien
    -0.79
    jshint
    -0.79
    POSITIVE LOGITS
     projected
    0.94
     Projection
    0.91
     projection
    0.90
     (
    0.89
     Projected
    0.81
     projecting
    0.80
    ？　
    0.79
     projections
    0.79
    Projection
    0.79
    ̠
    0.78
    Activations Density 0.022%

    No Known Activations