© 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. 99587
    Prev
    Next
    INDEX
    Explanations

    Malayalam, Ques, einen, quest, monoch, quantidade, whatsapp

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

    The neuron flags out-of-vocabulary or non-English tokens (foreign words or names) in the 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
     and
    -2.56
    jenigen
    -2.41
     theses
    -2.28
     mencionar
    -2.27
    mathrm
    -2.17
     moglie
    -2.16
     ponerse
    -2.14
    ႐
    -2.11
     Когда
    -2.09
     regresar
    -2.09
    POSITIVE LOGITS
    縯
    2.45
    峮
    2.33
    лока
    2.27
    [])
    2.25
     hommage
    2.16
     mélange
    2.13
    medriver
    2.13
     terke
    2.05
     &&
    2.05
     }),
    2.05
    Activations Density 0.001%

    No Known Activations