© 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. 73198
    Prev
    Next
    INDEX
    Explanations

    nature and natural environments

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

    The neuron strongly activates on occurrences of the word “nature” (including forms like “nature’s”)—i.e. explicit references to the natural world.

    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
    玙
    -3.00
    ৢ
    -2.08
    琍
    -2.08
    .")
    -2.06
    䊏
    -1.97
    gráfica
    -1.95
    ᷆
    -1.93
    馓
    -1.91
    䜣
    -1.91
    chuckles
    -1.90
    POSITIVE LOGITS
    .
    2.34
     You
    2.16
    </b>
    2.11
    <i>
    2.08
    </td>
    2.02
     -
    1.95
     «
    1.91
     want
    1.86
     The
    1.85
     do
    1.85
    Activations Density 0.011%

    No Known Activations