© 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. 69913
    Prev
    Next
    INDEX
    Explanations

    world war I entities

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

    The neuron primarily fires on numeric tokens (e.g. years, dates, counts, or other number strings).

    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
     Hitler
    -1.00
     Nazi
    -0.85
     USSR
    -0.77
    Hitler
    -0.74
     thập
    -0.74
     Lajos
    -0.73
     Nazis
    -0.73
     Polaroid
    -0.71
     WWII
    -0.69
    PROV
    -0.69
    POSITIVE LOGITS
     Woodrow
    0.82
     Coronel
    0.79
     Serbia
    0.79
     Zimmermann
    0.73
    statusText
    0.72
     Sarajevo
    0.71
    弩
    0.71
     gere
    0.71
    inputs
    0.71
     pria
    0.71
    Activations Density 0.029%

    No Known Activations