© 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. 112262
    Prev
    Next
    INDEX
    Explanations

    over time

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

    The neuron selectively fires on the semantically “key” content words (the main nouns, verbs, and adjectives) in each sentence.

    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
     veulent
    -1.90
    ܆
    -1.54
     chanteur
    -1.54
     trouver
    -1.46
    critti
    -1.45
     apprendre
    -1.44
     сала
    -1.42
     savent
    -1.41
     vermel
    -1.40
     choisir
    -1.40
    POSITIVE LOGITS
     that
    1.93
     what
    1.73
     three
    1.70
     five
    1.70
     eight
    1.64
     has
    1.63
     two
    1.63
     fourth
    1.60
     becomes
    1.59
     provide
    1.58
    Activations Density 0.141%

    No Known Activations