© 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. 34-GEMMASCOPE-RES-131K
    7. 63699
    Prev
    Next
    INDEX
    Explanations

    to make / made

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

    The neuron fires on bits of “how-to” instructional language—especially pronouns and connectives in DIY or recipe steps (e.g. “I,” “how,” “easy,” “to make,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     hilos
    -0.90
    atuan
    -0.88
    ytu
    -0.85
     projets
    -0.85
     incidencia
    -0.82
    }}(\
    -0.80
    НИ
    -0.79
     состава
    -0.79
    Setup
    -0.79
    デザイナー
    -0.79
    POSITIVE LOGITS
     made
    3.94
     making
    3.77
     make
    3.75
     membuat
    2.84
    Making
    2.81
     makes
    2.80
    making
    2.78
     만들
    2.58
    MAKING
    2.50
     Making
    2.45
    Activations Density 0.174%

    No Known Activations