© 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. 10-GEMMASCOPE-RES-131K
    7. 57635
    Prev
    Next
    INDEX
    Explanations

    time left in games

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

    The neuron activates on numeric expressions—particularly measurements of time, scores, or other decimal‐ and integer‐based figures.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    doesn
    -1.98
    those
    -1.95
    apparently
    -1.88
    excellent
    -1.80
    unfortunately
    -1.80
    simply
    -1.79
    several
    -1.77
    didn
    -1.76
    extremely
    -1.75
    そういえば
    -1.75
    POSITIVE LOGITS
     of
    1.77
     A
    1.66
     information
    1.56
     or
    1.45
     oraz
    1.43
     eine
    1.41
     focused
    1.38
     Having
    1.37
     time
    1.36
     If
    1.35
    Activations Density 0.004%

    No Known Activations