© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 2435
    Prev
    Next
    INDEX
    Explanations

    playoffs and finals

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

    The neuron strongly activates on words naming competitive match stages (e.g., “round,” “semifinals,” “quarterfinals,” “final”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     prokary
    0.77
     ৭০
    0.76
     самостоя
    0.76
     ἀ
    0.74
     ৪০
    0.72
     phd
    0.71
     ৪৫
    0.70
     vissa
    0.70
    
    0.69
     diesem
    0.68
    POSITIVE LOGITS
    finals
    0.70
    btnClear
    0.70
     Finals
    0.69
     lineup
    0.68
     semifinals
    0.68
    嘉宾
    0.68
     펼
    0.66
    гови
    0.65
     विकेट
    0.64
    playSound
    0.64
    Activations Density 0.004%

    No Known Activations