© 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-3-27B
    5. 53-GEMMASCOPE-2-RES-65K
    7. 6222
    Prev
    Next
    INDEX
    Explanations

    agreement and interjections

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

    conversational filler and polite interjections

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_53_width_65k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    =$(
    0.41
     Due
    0.41
    \\
    0.40
    由于
    0.39
    "],
    0.38
    Due
    0.38
     있으며
    0.38
     जिसमे
    0.37
     דבר
    0.37
    )="
    0.36
    POSITIVE LOGITS
     huh
    1.34
     eh
    1.05
    huh
    0.96
     yaar
    0.93
     sir
    0.88
     dear
    0.82
    sir
    0.80
     hmm
    0.79
     haha
    0.78
    啊
    0.77
    Activations Density 0.207%

    No Known Activations