© 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. 68670
    Prev
    Next
    INDEX
    Explanations

    non-english characters and markdown

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

    The neuron fires on the token “com,” i.e. occurrences of the “com” segment in Java/Android package or import statements.

    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
     entró
    -1.28
     trattano
    -1.16
     slike
    -1.13
    imédia
    -1.09
     both
    -1.09
    -*
    -1.08
    ':{'
    -1.08
    ToMany
    -1.08
     refiri
    -1.08
    ledad
    -1.07
    POSITIVE LOGITS
    1.07
    ฌ
    1.07
     попробу
    1.05
     electrica
    1.04
     пишет
    1.01
     этой
    0.98
    Dope
    0.97
    晤
    0.96
     positiva
    0.96
    ściej
    0.94
    Activations Density 0.001%
    No Known Activations