© 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. 10-GEMMASCOPE-RES-131K
    4. 110906
    Prev
    Next
    INDEX
    Explanations

    names like lucifer, michael, apost, exor

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

    The neuron fires on proper names, titles, and other named-entity tokens (e.g. character names, show/movie/game titles, genus names, etc.).

    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
     figliu
    -1.44
     italienischen
    -1.43
    </h5>
    -1.39
     objected
    -1.38
     includes
    -1.35
    </h3>
    -1.32
     consists
    -1.30
     rinnov
    -1.30
     adesi
    -1.26
    大概是
    -1.24
    POSITIVE LOGITS
     of
    1.52
     your
    1.45
    mathrm
    1.30
    Спа
    1.22
     Przed
    1.13
     not
    1.11
    ların
    1.10
    จาก
    1.08
     all
    1.07
    вість
    1.05
    Activations Density 0.007%

    No Known Activations