© 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. 20355
    Prev
    Next
    INDEX
    Explanations

    proper nouns and beginnings of terms

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

    The neuron fires on capitalized named entities—proper nouns such as people’s names or titled works.

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

    capitalized proper names and title words, especially named entities from works, people, organizations, and series.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    proper nouns, particularly names of people and places.

    oai_token-act-pair · claude-4-5-sonnetTriggered 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
     gambe
    -1.40
    ;%%
    -1.37
     vó
    -1.31
     zei
    -1.30
     With
    -1.29
    jaciół
    -1.25
     apporté
    -1.24
    ?,?,
    -1.23
     specialmente
    -1.23
    fetera
    -1.20
    POSITIVE LOGITS
    <bos>
    1.84
    筏
    1.26
    提前
    1.21
    すでに
    1.20
    納豆
    1.19
    袂
    1.18
     gradualmente
    1.18
    Musical
    1.17
    Música
    1.16
     agrícola
    1.15
    Activations Density 0.002%

    No Known Activations