© 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. 31492
    Prev
    Next
    INDEX
    Explanations

    set on fire

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

    The neuron fires on constructions describing something being set (or setting) “on fire” (i.e. phrases where an object is put or goes on 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
    bigcap
    -0.79
     мара
    -0.77
    hik
    -0.71
    ティスト
    -0.69
    hel
    -0.69
    勋
    -0.69
     CBT
    -0.69
    heard
    -0.68
    域
    -0.67
    dbd
    -0.67
    POSITIVE LOGITS
     fire
    2.52
     ablaze
    2.11
     flames
    1.86
     fires
    1.79
    FIRE
    1.73
    fire
    1.71
     burning
    1.70
     fuego
    1.66
     On
    1.66
     fuoco
    1.63
    Activations Density 0.025%

    No Known Activations