© 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. 22-GEMMASCOPE-RES-131K
    7. 12702
    Prev
    Next
    INDEX
    Explanations

    validity, unknown, diagnosis

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

    The neuron strongly activates on content‐bearing, domain‐specific terms (rare/technical nouns and vocabulary) rather than on common function words.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     repuesto
    -0.94
    gitter
    -0.93
     marchandises
    -0.93
    许多
    -0.92
    setzungen
    -0.89
     שלך
    -0.88
     réparer
    -0.85
    ății
    -0.85
     mať
    -0.84
     prévenir
    -0.84
    POSITIVE LOGITS
    lové
    0.93
     original
    0.85
    全力
    0.83
    に行ってきました
    0.82
     initial
    0.82
     without
    0.80
     sentimental
    0.79
     cerve
    0.78
    Inicial
    0.78
    gypten
    0.77
    Activations Density 0.075%

    No Known Activations