© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 124101
    Prev
    Next
    INDEX
    Explanations

    by or with phrase connector

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

    This neuron fires on relational prepositions (especially “by,” “with,” and “in”) that introduce how or by what means something is done.

    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
     they
    -1.62
    colazione
    -1.46
    lepaskan
    -1.40
     such
    -1.39
     bedoeld
    -1.34
    alamualaikum
    -1.32
     overeen
    -1.31
    ‽
    -1.31
    由于
    -1.30
     that
    -1.29
    POSITIVE LOGITS
    }
    1.49
    g
    1.48
    b
    1.47
     a
    1.41
    </h4>
    1.41
    ga
    1.38
    的金
    1.37
     journaliste
    1.34
     ANOTHER
    1.32
    </h3>
    1.31
    Activations Density 0.068%

    No Known Activations