© 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. 54123
    Prev
    Next
    INDEX
    Explanations

    magnetic coils and fields

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

    The neuron is specialized for detecting numeric tokens, hyphens, and numbered section/item references in technical or patent‐style 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_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
    верка
    -0.83
     trop
    -0.78
     térmico
    -0.76
    Vectors
    -0.74
     Chad
    -0.74
     brady
    -0.73
     cele
    -0.72
    Dil
    -0.71
     transformación
    -0.71
     Gibb
    -0.69
    POSITIVE LOGITS
     coil
    1.68
     coils
    1.68
    coil
    1.49
     Coil
    1.35
    Coil
    1.26
     magnet
    1.09
    magnet
    1.07
     Bitter
    1.02
     magnets
    1.01
     windings
    0.96
    Activations Density 0.063%

    No Known Activations