© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 74266
    Prev
    Next
    INDEX
    Explanations

    animation, aspiration, historical British, legal trust, API, oxidases

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

    The neuron is highly sensitive to the stems of words formed with the English “-ation” suffix (e.g. “aspir-” in “aspiration”) or its Russian “-ция” counterpart (e.g. “анима-” in “анимация”), i.e. it fires on the root portion of nouns ending in “-ation/-ция.”

    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
    "
    -2.56
    0
    -2.45
    If
    -2.42
     and
    -2.11
     powinien
    -2.09
     If
    -2.09
    ↵↵
    -1.99
    			
    -1.98
    avía
    -1.95
    といけない
    -1.91
    POSITIVE LOGITS
    ის
    3.17
    屃
    2.42
    幛
    2.41
    ا
    2.30
    蚨
    2.27
     We
    2.23
    e
    2.22
    as
    2.22
    ⃚
    2.16
    蠃
    2.16
    Activations Density 0.000%

    No Known Activations