© 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. 115929
    Prev
    Next
    INDEX
    Explanations

    under attack or strike

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

    The neuron fires on words signaling violent or aggressive actions—especially attacks, assaults, strikes, or being “under fire.”

    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
    Appellee
    -0.78
    Excessive
    -0.72
     ибо
    -0.72
    ItemName
    -0.72
    octobre
    -0.70
    疫
    -0.69
     Zeugen
    -0.69
    leč
    -0.68
    olat
    -0.68
    olato
    -0.67
    POSITIVE LOGITS
     attack
    6.00
     attacks
    5.59
    attack
    4.81
     attacked
    4.66
    Attack
    4.34
     Attack
    4.31
    attacks
    4.16
     ATTACK
    4.03
     Attacks
    3.97
     ataque
    3.92
    Activations Density 0.096%

    No Known Activations