© 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. 68524
    Prev
    Next
    INDEX
    Explanations

    dates, questions, and cases

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

    The neuron strongly activates on the comma that separates exception types in a Java `throws` clause.

    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
    子は
    -1.13
    子を
    -1.05
     matériel
    -1.05
    脖
    -1.04
    vereign
    -1.02
     each
    -1.01
    резать
    -0.98
    ๔
    -0.98
    tely
    -0.96
    そして
    -0.94
    POSITIVE LOGITS
     ebenfalls
    1.09
    esgue
    1.05
    Dette
    1.04
     afore
    1.02
    xido
    0.99
     aforementioned
    0.98
     cajones
    0.93
    いかがでしたか
    0.91
    クトル
    0.91
    utilizzo
    0.90
    Activations Density 0.000%

    No Known Activations