© 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. 76320
    Prev
    Next
    INDEX
    Explanations

    binding function calls

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

    The neuron consistently activates on number literals (any token containing 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
     betalen
    -0.84
     trekken
    -0.78
     rijden
    -0.77
     volverá
    -0.75
     bereiken
    -0.75
     apoyar
    -0.75
     zoeken
    -0.74
    Carreira
    -0.72
     genieten
    -0.72
     kiezen
    -0.72
    POSITIVE LOGITS
    bind
    0.94
     however
    0.81
    ics
    0.78
     statements
    0.77
     экскур
    0.77
     dessa
    0.77
     bind
    0.75
    trigger
    0.72
     learnt
    0.71
     especie
    0.71
    Activations Density 0.023%

    No Known Activations