© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 68226
    Prev
    Next
    INDEX
    Explanations

    future action or consequence

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

    The neuron strongly activates on bold, future‐oriented promise verbs in promotional or hype contexts (e.g. “will blow you away,” “will make your day,” “will keep you entertained”).

    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
     it
    -2.06
     creando
    -1.40
    了很多
    -1.40
     then
    -1.39
     atuação
    -1.38
    については
    -1.32
     who
    -1.30
     myriad
    -1.27
    um
    -1.27
     how
    -1.27
    POSITIVE LOGITS
    </tfoot>
    1.60
    ׇ
    1.58
     {};
    
    1.56
     and
    1.53
     *}
    1.48
     restaurantes
    1.45
    டிய
    1.45
    洳
    1.44
     "}
    1.43
    )("
    1.43
    Activations Density 0.035%

    No Known Activations