© 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. 32927
    Prev
    Next
    INDEX
    Explanations

    known prior art

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

    The neuron fires strongly on the key descriptive content words—mostly salient nouns and adjectives that convey the main features or attributes in a listing or article.

    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
    Bibliografie
    -1.13
    স
    -1.09
     Ordem
    -1.04
     hará
    -1.02
    reuters
    -0.98
     Sha
    -0.98
    НОВ
    -0.97
    са
    -0.95
    ができます
    -0.94
    有點
    -0.93
    POSITIVE LOGITS
     brasileiros
    1.35
     preciosas
    1.33
     meninos
    1.33
     coloridas
    1.31
     katastro
    1.23
     patienter
    1.23
     ägg
    1.19
     decorativas
    1.18
     män
    1.16
     ekolog
    1.16
    Activations Density 0.372%

    No Known Activations