© 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. 100012
    Prev
    Next
    INDEX
    Explanations

    buoyancy, flotation, and settling

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

    The neuron fires on technical terms for density‐based separation and support—e.g. centrifugation/ultracentrifugation, sedimentation, buoyancy, flotation, and related numeric measurements.

    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
    sandalias
    -1.38
    けません
    -1.29
    nF
    -1.27
     brauchen
    -1.26
    ettare
    -1.23
    骯
    -1.23
     verschiedene
    -1.22
     Vorsitzende
    -1.21
     شدن
    -1.20
     текста
    -1.18
    POSITIVE LOGITS
     instead
    1.77
    instead
    1.63
    ’
    1.39
    chiato
    1.34
    𓋼
    1.34
    之上
    1.32
    ]
    1.31
     hobbies
    1.29
    じゃなくて
    1.29
     rather
    1.27
    Activations Density 0.041%

    No Known Activations