© 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. 110880
    Prev
    Next
    INDEX
    Explanations

    halftime and overtime

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

    The neuron strongly activates on terms marking numeric or temporal events in sports reporting (e.g. half-time, second-half, stroke of the half, point, two goals, etc.).

    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
    lière
    -0.82
    ⠁
    -0.81
     mesi
    -0.78
     πρω
    -0.76
     nächste
    -0.70
     semanas
    -0.70
    SMP
    -0.69
    ೨
    -0.68
    annin
    -0.68
    andr
    -0.68
    POSITIVE LOGITS
     halftime
    2.47
     overtime
    1.89
     extra
    1.87
    extra
    1.59
     regulation
    1.53
    Extra
    1.52
     minute
    1.45
    EXTRA
    1.41
     half
    1.40
    Minute
    1.38
    Activations Density 0.033%

    No Known Activations