© 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. 22-GEMMASCOPE-RES-131K
    7. 67116
    Prev
    Next
    INDEX
    Explanations

    This invention relating to

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

    The neuron activates on numeric literals (e.g. multi‐digit numbers) in technical or code‐like text.

    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
     if
    -2.05
     (
    -1.76
     without
    -1.74
     Drž
    -1.67
     because
    -1.63
     had
    -1.59
     new
    -1.58
    子は
    -1.57
     whose
    -1.57
     two
    -1.55
    POSITIVE LOGITS
     essas
    1.89
     Dinas
    1.63
     molte
    1.57
     horrid
    1.57
     stalwart
    1.53
     alcune
    1.52
     REALLY
    1.49
    Continu
    1.48
     insanely
    1.43
     acht
    1.42
    Activations Density 0.000%

    No Known Activations