© 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. 67398
    Prev
    Next
    INDEX
    Explanations

    starts of compound words or specific structures

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

    The neuron remains silent on all the example text—it does not reliably fire on any particular token or pattern.

    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
    itions
    -1.00
     beiden
    -0.98
    kish
    -0.98
    debe
    -0.97
     sucht
    -0.96
     दोनों
    -0.93
    irie
    -0.91
    atzes
    -0.91
    "',
    -0.91
     pertama
    -0.90
    POSITIVE LOGITS
     also
    1.11
     another
    0.93
    有一個
    0.91
    6
    0.88
    ことがある
    0.88
     خوبه
    0.86
    可能會
    0.85
    さんも
    0.85
     предыду
    0.85
     fave
    0.84
    Activations Density 0.000%

    No Known Activations