© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 10297
    Prev
    Next
    INDEX
    Explanations

    introduced again

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

    The neuron highlights high‐information, content‐bearing words (salient keywords) in a text.

    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
     for
    -2.31
     because
    -2.06
     of
    -2.05
     this
    -1.60
     kelahiran
    -1.55
     delapan
    -1.54
    ſelves
    -1.51
     eight
    -1.49
     you
    -1.46
     buatan
    -1.45
    POSITIVE LOGITS
     again
    1.48
    </em>
    1.30
    fichier
    1.27
     опять
    1.27
     Brings
    1.27
     quí
    1.26
    摟
    1.24
    moteur
    1.23
    catalogo
    1.23
    Again
    1.23
    Activations Density 0.206%

    No Known Activations