© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 130233
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on words referring to death, dying, or other fatal outcomes.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    death and doom

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     conejo
    -1.18
    dịch
    -1.15
     umě
    -1.05
     Songtext
    -1.02
     Kunststoff
    -1.02
    ánicos
    -1.02
     obf
    -1.01
     stdClass
    -1.00
     peindre
    -0.99
     Icelandic
    -0.98
    POSITIVE LOGITS
     death
    1.54
     died
    1.30
     deaths
    1.29
     die
    1.15
     demise
    1.12
     либо
    1.09
     either
    1.08
    death
    1.03
     suicide
    1.00
     fate
    0.99
    Activations Density 0.033%

    No Known Activations