© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 128704
    Prev
    Next
    INDEX
    Explanations

    reported statements and findings

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

    The neuron spotlights mentions of information sources—e.g. publications, papers, reports, journals, officials, and news outlets.

    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
    Dichloropropene
    -1.66
     Institutionen
    -1.49
     a
    -1.40
     ambos
    -1.38
     uneasiness
    -1.37
    櫛
    -1.34
     Familie
    -1.33
    })),
    -1.32
    licitud
    -1.31
     beiden
    -1.28
    POSITIVE LOGITS
     also
    3.16
     said
    1.91
     further
    1.77
    also
    1.64
    了嗎
    1.54
    是因為
    1.45
     again
    1.41
     lupo
    1.37
     compared
    1.36
     wiem
    1.34
    Activations Density 0.041%

    No Known Activations