© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 15544
    Prev
    Next
    INDEX
    Explanations

    toasting and roasting food

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

    The neuron strongly activates on terms naming dry‐heat cooking or processing methods (e.g. roasting, toasting, torrefaction) and related items.

    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
    ENOMEM
    -1.80
    irtschaft
    -1.68
    拊
    -1.67
     chisel
    -1.60
    橼
    -1.60
    "?
    -1.54
    婳
    -1.51
    都已经
    -1.48
    平时
    -1.48
    IA
    -1.46
    POSITIVE LOGITS
     –
    1.74
    ’
    1.70
    ные
    1.48
    ↵
    1.47
    ificación
    1.45
    Médaille
    1.45
    3
    1.45
    ．．．
    1.44
     ‘
    1.43
    šana
    1.43
    Activations Density 0.030%

    No Known Activations