© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 91471
    Prev
    Next
    INDEX
    Explanations

    numbers and punctuation

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

    The neuron detects the curly apostrophe character used in English contractions (e.g. “we’re,” “isn’t,” “what’s,” “I’m,” etc.).

    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
     extraordinaire
    -0.76
    발
    -0.75
    eef
    -0.73
     giorni
    -0.73
     fisica
    -0.72
    iD
    -0.71
    джин
    -0.71
    俗
    -0.69
     Oriented
    -0.69
     vloer
    -0.69
    POSITIVE LOGITS
    Related
    0.93
    Exclusive
    0.89
     view
    0.81
    view
    0.81
    rapid
    0.77
    olata
    0.76
     закри
    0.75
    COLL
    0.75
    expanding
    0.74
    cleared
    0.74
    Activations Density 0.059%

    No Known Activations