© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 104513
    Prev
    Next
    INDEX
    Explanations

    identifiers and properties

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

    The neuron activates on numeric tokens—particularly digits and number literals (e.g. IDs, constants) in code.

    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
    pictured
    -0.84
    笺
    -0.83
     accepts
    -0.80
     "'");
    -0.77
    ]}\
    -0.77
     Dangerous
    -0.77
     }{@
    -0.76
     impra
    -0.75
    fis
    -0.75
    Touchable
    -0.75
    POSITIVE LOGITS
     Baru
    0.88
    ladel
    0.88
     recent
    0.86
     montagna
    0.85
     already
    0.84
     уже
    0.84
     planned
    0.83
     new
    0.83
     Calcium
    0.83
    。"
    0.83
    Activations Density 0.007%

    No Known Activations