© 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. 34-GEMMASCOPE-RES-131K
    4. 89772
    Prev
    Next
    INDEX
    Explanations

    professional titles and roles

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

    The neuron activates on capitalized tokens that are part of named entities or formal titles—e.g. people’s names, official positions, organizations, and other proper nouns.

    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
     produisons
    -1.20
     mittlere
    -1.17
    ですし
    -0.89
     originalen
    -0.88
     which
    -0.86
    Ayuda
    -0.85
    swire
    -0.85
    OBJETIVOS
    -0.85
    </b>
    -0.84
    Pessoa
    -0.84
    POSITIVE LOGITS
     colleague
    1.41
     veteran
    1.39
     heavyweight
    1.38
     guru
    1.34
     superstar
    1.30
     sensation
    1.26
     newcomer
    1.26
     businessman
    1.21
     friend
    1.20
     researcher
    1.20
    Activations Density 0.216%

    No Known Activations