© 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. 34-GEMMASCOPE-RES-131K
    7. 94039
    Prev
    Next
    INDEX
    Explanations

    questions followed by numbers or letters

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

    This neuron primarily detects standalone numeric tokens (e.g. question or choice numbers and other digits).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     sueños
    -0.99
    ọt
    -0.93
    ruguay
    -0.91
     botines
    -0.89
     peinados
    -0.89
     ulike
    -0.86
     offent
    -0.85
    Сылтамалар
    -0.84
    ujemy
    -0.84
     acoged
    -0.83
    POSITIVE LOGITS
     questions
    3.16
     question
    3.03
     Questions
    2.30
    Questions
    2.23
     Question
    2.17
    questions
    2.11
     what
    2.05
    question
    1.93
     how
    1.90
    Question
    1.89
    Activations Density 0.008%

    No Known Activations