© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 78974
    Prev
    Next
    INDEX
    Explanations

    assassination and related words

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

    The neuron specializes in detecting explicit date and time markers—especially numeric year tokens and closely associated temporal words in historical contexts.

    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
     Gesells
    -0.85
     erb
    -0.79
    MENAFN
    -0.79
     góp
    -0.78
     znaczy
    -0.75
     equipping
    -0.75
     inflicting
    -0.73
     belligerent
    -0.73
     cắm
    -0.72
     Crush
    -0.70
    POSITIVE LOGITS
     assassination
    4.19
     assassin
    3.91
     assassinated
    3.48
    assassin
    2.98
     Assassin
    2.80
    Assassin
    2.78
     assassins
    2.75
    assination
    2.17
     assass
    2.14
     murder
    1.70
    Activations Density 0.044%

    No Known Activations