© 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. 82618
    Prev
    Next
    INDEX
    Explanations

    with pronouns/articles

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

    The neuron strongly activates on occurrences of the preposition “with.”

    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
    gnon
    -0.98
    もあって
    -0.94
    entr
    -0.93
    があります
    -0.93
    見せ
    -0.91
    ce
    -0.89
    actually
    -0.88
     exchanges
    -0.88
     been
    -0.87
     relationships
    -0.87
    POSITIVE LOGITS
     us
    1.77
     them
    1.60
    standing
    1.13
     him
    1.13
     the
    1.12
     impunity
    1.09
    secuted
    1.09
     nhau
    1.00
    ❯
    0.97
    に乗って
    0.97
    Activations Density 0.025%

    No Known Activations