© 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. 446
    Prev
    Next
    INDEX
    Explanations

    keeping things going or intact

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

    keep in mind

    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
     segera
    0.47
    馬上
    0.46
    already
    0.42
     сразу
    0.41
    立马
    0.40
     Already
    0.40
    立即
    0.40
    早
    0.40
    quickly
    0.40
    立刻
    0.39
    POSITIVE LOGITS
     intact
    1.33
     alive
    1.20
     unchanged
    1.13
     afloat
    1.09
    alive
    1.06
     abreast
    0.98
     unaltered
    0.94
    不变
    0.90
     steadfast
    0.90
     vigilant
    0.89
    Activations Density 0.381%

    No Known Activations