© 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. 104762
    Prev
    Next
    INDEX
    Explanations

    classifications, data, observed

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

    The neuron strongly activates on quantitative and measurement tokens—numbers, units, counts, and other numerical data often found in scientific‐paper abstracts.

    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
    𝑞
    -1.54
     السياس
    -1.51
     APPELLANT
    -1.47
     xícara
    -1.45
     condiv
    -1.40
     Bezir
    -1.39
     how
    -1.38
     Songtext
    -1.38
     sogenannten
    -1.36
    𝜌
    -1.36
    POSITIVE LOGITS
    complished
    1.74
    vaila
    1.66
     udang
    1.63
     सत्यापित
    1.52
    COLORS
    1.52
     нещо
    1.46
    hoka
    1.42
     saveurs
    1.38
     BoxFit
    1.37
    Público
    1.37
    Activations Density 0.084%

    No Known Activations