© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 47766
    Prev
    Next
    INDEX
    Explanations

    SPDX licenses

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

    The neuron is a subword detector that activates on the character sequence “Schul.”

    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
    ATION
    -2.28
     ponen
    -2.25
     hinder
    -2.22
     grupo
    -2.22
    untamiento
    -2.19
     今
    -2.16
    所有人都
    -2.13
    ที่มี
    -2.11
     illuminating
    -2.06
     представители
    -2.06
    POSITIVE LOGITS
    0
    3.14
    ly
    2.94
     daer
    2.83
    鄚
    2.64
     esat
    2.52
    焿
    2.52
     kebaya
    2.42
     moter
    2.42
     たん
    2.39
     sara
    2.38
    Activations Density 0.003%

    No Known Activations