© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 99108
    Prev
    Next
    INDEX
    Explanations

    Pompeii, cities, eruption, burnt

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

    The neuron activates almost exclusively on numeric tokens or digit‐containing strings (years, measurements, counts, 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
     divides
    -0.75
    矛盾
    -0.73
     old
    -0.73
     discarded
    -0.72
     DEFENSE
    -0.70
     conflicts
    -0.70
     noticeable
    -0.68
    Jug
    -0.68
    meat
    -0.68
     utilized
    -0.67
    POSITIVE LOGITS
     Pompeii
    1.40
    Pom
    1.07
     eruption
    0.89
     buried
    0.88
    rimento
    0.88
     pompe
    0.86
     Pompe
    0.84
     Cities
    0.83
     města
    0.82
    ruzione
    0.80
    Activations Density 0.008%

    No Known Activations