© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 112010
    Prev
    Next
    INDEX
    Explanations

    building structures or units

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

    The neuron fires on text describing real-time strategy game mechanics—especially verbs and nouns around gathering resources and building units.

    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
    ceptable
    -1.63
     {}));
    -1.43
    nachron
    -1.41
     Memoriam
    -1.38
    dirond
    -1.35
    listair
    -1.34
    parently
    -1.33
     />}
    -1.32
    cketed
    -1.31
    けば
    -1.30
    POSITIVE LOGITS
     all
    1.49
     will
    1.41
     like
    1.40
     on
    1.30
     it
    1.28
    <b>
    1.23
     not
    1.21
    _
    1.20
     That
    1.17
    #
    1.15
    Activations Density 0.038%

    No Known Activations