© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 87385
    Prev
    Next
    INDEX
    Explanations

    concrete structure, leaked oil, honey bear

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

    The neuron detects descriptive adjectives or nouns that specify concrete physical attributes—shape, color, texture, size—of objects or scenes.

    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
     to
    -1.59
    presumably
    -1.34
    Familien
    -1.30
    velocityX
    -1.30
    鵟
    -1.27
    primarily
    -1.26
     aient
    -1.21
     vont
    -1.20
    ܆
    -1.20
    suddenly
    -1.20
    POSITIVE LOGITS
     by
    1.65
    They
    1.27
     on
    1.21
     musées
    1.16
    Each
    1.15
    from
    1.12
     perguntou
    1.10
    If
    1.10
    makaian
    1.07
    长江
    1.06
    Activations Density 0.148%

    No Known Activations