© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 7236
    Prev
    Next
    INDEX
    Explanations

    phrases starting with and

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

    The neuron flags informal discourse markers and filler/interjection phrases (e.g. “look,” “hey,” “oh yes,” “let’s face it,” “good for,” “for example,” “at least a few,” “be sure to,” “hell,” 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_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
     however
    -1.80
     However
    -1.75
    However
    -1.69
     però
    -1.57
     však
    -1.55
     Many
    -1.50
     A
    -1.46
     jedoch
    -1.44
     Also
    -1.39
    Some
    -1.37
    POSITIVE LOGITS
     trajets
    1.67
     Reſ
    1.61
     others
    1.57
     filtres
    1.51
     другие
    1.51
     other
    1.46
     falda
    1.46
    
    1.41
     toalla
    1.39
     différents
    1.38
    Activations Density 0.055%

    No Known Activations