© 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. 66125
    Prev
    Next
    INDEX
    Explanations

    historical figures and periods

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

    The neuron strongly activates on historical or encyclopedic content words—especially proper names, dates, and places—i.e. biographical or chronologically anchored information.

    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
     Рим
    -0.78
     toxic
    -0.77
     Ptolemy
    -0.76
    UNTAS
    -0.68
     medieval
    -0.68
    жки
    -0.68
    三国
    -0.68
    MinValue
    -0.66
     Romans
    -0.66
    urches
    -0.66
    POSITIVE LOGITS
     Assyrian
    0.97
     Herodotus
    0.84
     sixth
    0.84
     VI
    0.80
     Alc
    0.78
     Thales
    0.77
     DLC
    0.76
     Webb
    0.76
     ECU
    0.75
    YTE
    0.75
    Activations Density 0.028%

    No Known Activations