© 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. 100947
    Prev
    Next
    INDEX
    Explanations

    actions, shares, results

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

    The neuron fires on short French tokens (especially common function words like “en,” “ai,” “rien,” “pour,” etc.), effectively detecting French 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
    after
    -1.80
    from
    -1.67
    out
    -1.55
    e
    -1.53
    the
    -1.51
    좆
    -1.51
    掲示
    -1.48
    褙
    -1.48
    ENGTH
    -1.47
    你不是
    -1.47
    POSITIVE LOGITS
     which
    1.59
    AdvEx
    1.56
     klokken
    1.52
     včetně
    1.49
    providedIn
    1.48
     omtrent
    1.48
     danske
    1.45
     klasse
    1.37
    ações
    1.36
     eivät
    1.34
    Activations Density 0.001%

    No Known Activations