© 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. 57532
    Prev
    Next
    INDEX
    Explanations

    questions and answers

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

    The neuron detects occurrences of question labels or markers (e.g. “Q:” and the word “question”) in the text.

    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
     "
    -1.71
     be
    -1.66
     You
    -1.63
     nicht
    -1.61
     algunos
    -1.59
     Lovely
    -1.59
     he
    -1.59
    ]
    -1.58
    </sup>
    -1.55
    bekistan
    -1.55
    POSITIVE LOGITS
    脸上的
    1.86
    ✱
    1.72
     @"/
    1.69
     couvrir
    1.66
     ausreich
    1.63
     Signalez
    1.63
     verbess
    1.63
     TIPO
    1.61
    XICO
    1.59
    `/
    1.58
    Activations Density 0.237%

    No Known Activations