© 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. 22-GEMMASCOPE-RES-131K
    7. 73552
    Prev
    Next
    INDEX
    Explanations

    deep down

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

    This neuron reliably fires on occurrences of the adjective “deep.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     he
    -2.67
     digitalen
    -2.59
     reinen
    -2.50
     passenden
    -2.48
    >
    -2.42
     älteren
    -2.41
    霎
    -2.41
     ก็
    -2.41
     kleineren
    -2.41
     solchen
    -2.39
    POSITIVE LOGITS
    s
    3.27
    𖥦
    2.64
    ଝ
    2.61
    ୌ
    2.59
    鋮
    2.55
    After
    2.45
     éditoriale
    2.42
     стоковые
    2.34
     tradisional
    2.31
     lakini
    2.31
    Activations Density 0.012%

    No Known Activations