© 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. 50289
    Prev
    Next
    INDEX
    Explanations

    rotating cutting parts

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

    This neuron fires on technical patent‐style descriptions, particularly listing machinery components or structural elements of an invention.

    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
    ٱ
    -0.87
     Assays
    -0.76
    housse
    -0.75
    isNotBlank
    -0.71
    Anthony
    -0.68
     မ
    -0.68
     accelerometer
    -0.68
    OutOfBounds
    -0.68
     autofocus
    -0.68
    terna
    -0.67
    POSITIVE LOGITS
     blades
    1.59
     knives
    1.19
     blade
    1.17
    blades
    1.00
     Knives
    0.98
     Blades
    0.97
     plates
    0.97
    Kni
    0.93
    blade
    0.93
     rotor
    0.92
    Activations Density 0.106%

    No Known Activations