© 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. 73344
    Prev
    Next
    INDEX
    Explanations

    researching

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

    The neuron detects mentions of doing research, i.e. occurrences of “researching” or “researched.”

    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
    </h2>
    -2.38
    ),
    -2.22
    ра
    -2.14
    '>
    -2.06
    ”,
    -1.99
    </em>
    -1.99
    ).
    -1.96
    ">
    -1.93
    teryx
    -1.87
    </h3>
    -1.85
    POSITIVE LOGITS
    ar
    2.47
    茈
    2.17
    瘓
    2.09
    〤
    1.91
    垻
    1.86
    衚
    1.84
    蓴
    1.84
     كيفية
    1.82
     hvor
    1.81
    彼らは
    1.80
    Activations Density 0.005%

    No Known Activations