© 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. 30596
    Prev
    Next
    INDEX
    Explanations

    method option destroy

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

    The neuron strongly activates on the title line of a StackOverflow post (the question’s header), picking up the main words of the question’s title.

    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
    uais
    -0.74
    greja
    -0.70
    ciare
    -0.70
    czes
    -0.70
    ieniem
    -0.69
     рынка
    -0.68
    rature
    -0.67
    CHOR
    -0.67
    рах
    -0.67
    IDTH
    -0.64
    POSITIVE LOGITS
     destroy
    1.11
    destroy
    1.09
     destroys
    0.91
    option
    0.88
    MethodInfo
    0.86
     option
    0.84
     options
    0.83
     destroyed
    0.81
     destroying
    0.81
    refresh
    0.81
    Activations Density 0.034%

    No Known Activations