© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 103686
    Prev
    Next
    INDEX
    Explanations

    Jaguar and

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

    The neuron strongly activates on standalone numeric tokens (e.g. years, quantities, model numbers).

    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
     Goul
    -0.76
     Uranus
    -0.76
    atius
    -0.74
    勃
    -0.74
     antioxid
    -0.73
     Tann
    -0.71
    ínico
    -0.71
    舶
    -0.71
     Startups
    -0.70
    hObject
    -0.69
    POSITIVE LOGITS
     Jaguar
    1.42
    Jaguar
    1.17
     British
    1.16
     Britain
    1.04
    British
    0.96
     британ
    0.94
     jaguar
    0.94
     Jaguars
    0.93
    jaguar
    0.87
    Britain
    0.84
    Activations Density 0.002%

    No Known Activations