© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 34665
    Prev
    Next
    INDEX
    Explanations

    conditional Portuguese and Spanish

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

    The neuron activates on very high‐frequency Portuguese (and some Spanish) function words and stopwords—e.g. “que,” “não,” “você,” “se,” “cuando,” etc.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     ayude
    -1.34
     llegue
    -1.19
     conoz
    -1.19
     lleve
    -1.16
     desee
    -1.16
     comien
    -1.10
     are
    -1.09
     parezca
    -1.09
     enví
    -1.07
     is
    -0.99
    POSITIVE LOGITS
     be
    3.91
     بتوان
    2.22
     legyen
    2.19
     باشد
    1.88
     будь
    1.86
     olsun
    1.69
    be
    1.38
     داشته
    1.37
     Be
    1.35
     soient
    1.34
    Activations Density 0.011%

    No Known Activations