© 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. 22-GEMMASCOPE-RES-131K
    7. 9649
    Prev
    Next
    INDEX
    Explanations

    PK Fc Q Hz Gain

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

    The neuron fires on numeric literal tokens, especially those denoting configuration parameters such as coordinates, sizes, rates, or other numeric settings.

    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
     their
    -1.41
     entonces
    -1.25
     just
    -1.23
     następnie
    -1.20
    ╗
    -1.16
    २
    -1.16
     his
    -1.15
     के
    -1.12
    綜合報導
    -1.09
     then
    -1.08
    POSITIVE LOGITS
    還可以
    1.34
     again
    1.26
    Materiaal
    1.23
     refirió
    1.23
     farther
    1.21
     хоче
    1.20
    Prijs
    1.20
    ܜ
    1.19
    Materiál
    1.19
    again
    1.19
    Activations Density 0.002%

    No Known Activations