© 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. 22-GEMMASCOPE-RES-131K
    7. 54448
    Prev
    Next
    INDEX
    Explanations

    Apollo, Thank, Oc, Q followed by punctuation/keywords

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

    The neuron activates on numeric literals in code (integer 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_22/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
     to
    -1.38
     berge
    -1.27
     kabar
    -1.17
     ¤
    -1.17
     décès
    -1.16
     gouverneur
    -1.15
     réalisateur
    -1.14
     lavabo
    -1.13
     blusa
    -1.13
     charmant
    -1.12
    POSITIVE LOGITS
     из
    0.98
     L
    0.97
     time
    0.95
     J
    0.93
    主流
    0.92
    𝐱
    0.91
     among
    0.90
    IFICACIONES
    0.90
     positive
    0.89
     T
    0.89
    Activations Density 0.001%

    No Known Activations