© 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. 83246
    Prev
    Next
    INDEX
    Explanations

    competition eliminations

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

    The neuron flags words that set up or refer to the timing or sequence of a major event (e.g. “after,” “finale,” “Tonight’s final,” “performance,” “at”).

    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
    Rusia
    -0.84
    PrototypeOf
    -0.81
    мерть
    -0.79
    COMPLETE
    -0.76
     boycott
    -0.76
     نداره
    -0.74
     títulos
    -0.73
    forhold
    -0.72
     olvidar
    -0.72
    løs
    -0.72
    POSITIVE LOGITS
     elimination
    1.48
     week
    1.23
     elimin
    1.16
    Elimin
    1.14
     Elimination
    1.10
     expuls
    1.07
    elimin
    1.07
     Elimin
    1.06
     voting
    1.05
     eliminated
    1.02
    Activations Density 0.010%

    No Known Activations