© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 40437
    Prev
    Next
    INDEX
    Explanations

    name initialisms abbreviations phrases

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

    The neuron fires on relatively long, low-frequency content tokens—e.g. proper names, technical terms, or other uncommon multi-syllable words.

    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
     it
    -1.96
     quella
    -1.45
     that
    -1.44
     los
    -1.40
     you
    -1.40
     hata
    -1.34
     quello
    -1.31
     akik
    -1.31
     dragón
    -1.30
    😜
    -1.30
    POSITIVE LOGITS
     demasi
    1.51
     листопада
    1.37
     パーティー
    1.35
    しているので
    1.34
     октября
    1.34
     oznacza
    1.27
    </h1>
    1.26
    guigu
    1.26
     ему
    1.24
    ၞ
    1.24
    Activations Density 0.160%

    No Known Activations