© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 39707
    Prev
    Next
    INDEX
    Explanations

    Spanish adverbs of manner

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

    The neuron primarily detects numeric literals (standalone numbers) in code or markup.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    (
    -2.38
    c
    -1.98
    //
    -1.96
     rapides
    -1.92
    To
    -1.87
    K
    -1.77
    e
    -1.73
    "
    -1.70
    On
    -1.66
    The
    -1.64
    POSITIVE LOGITS
     colocando
    1.88
     téc
    1.71
     frecuentemente
    1.69
     aparentemente
    1.68
    ieR
    1.66
    apj
    1.66
     posiblemente
    1.66
     básicamente
    1.66
     praticamente
    1.64
     generando
    1.62
    Activations Density 0.015%

    No Known Activations