© 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. 10-GEMMASCOPE-RES-131K
    7. 58788
    Prev
    Next
    INDEX
    Explanations

    phrases following specific words

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

    This neuron detects tokens involved in obituary‐style biographical details—especially mentions of death dates, circumstances of passing, ages, and related family/home context.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     céle
    -1.65
    暗い
    -1.48
    tizens
    -1.46
    Ꮴ
    -1.45
     estrena
    -1.45
    ⁶
    -1.45
    ܞ
    -1.43
     autorización
    -1.38
     suivent
    -1.34
     unrivalled
    -1.34
    POSITIVE LOGITS
     with
    1.70
     and
    1.70
     at
    1.56
     had
    1.55
     connected
    1.44
     took
    1.41
     started
    1.35
     began
    1.34
     did
    1.34
    ？』
    1.34
    Activations Density 0.012%

    No Known Activations