© 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. 43636
    Prev
    Next
    INDEX
    Explanations

    US states and Canadian places

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

    The neuron fires on factual “info‐box” style tokens—proper names, places, dates, measurements, and other discrete data points.

    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
     toques
    -1.41
    r
    -1.15
     valide
    -1.12
    }
    -1.09
     неё
    -1.09
     tumours
    -1.07
    しております
    -1.07
    </sub>
    -1.05
     neighbourhoods
    -1.05
    アレル
    -1.03
    POSITIVE LOGITS
    지
    1.33
     jednej
    1.30
    ১
    1.26
     améli
    1.22
    że
    1.18
    pax
    1.17
    TURE
    1.13
     InputMaybe
    1.11
    ম
    1.11
     jedno
    1.09
    Activations Density 0.277%

    No Known Activations