© 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. 15367
    Prev
    Next
    INDEX
    Explanations

    look at or think about

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

    The neuron responds strongly to occurrences of the verb “look,” especially in reflective phrases like “when I look.”

    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
     was
    -1.63
    *
    -1.58
    '
    -1.45
    }$
    -1.45
    .*
    -1.45
     who
    -1.36
     wasn
    -1.29
     didn
    -1.29
    .’
    -1.27
    —
    -1.24
    POSITIVE LOGITS
     různých
    1.66
     nových
    1.42
     beschrieben
    1.41
     vardır
    1.37
    daad
    1.34
    名字
    1.30
     různ
    1.29
    比如
    1.27
     things
    1.27
    hålla
    1.26
    Activations Density 0.116%

    No Known Activations