© 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-3-27B
    5. 31-GEMMASCOPE-2-RES-262K
    7. 81433
    Prev
    Next
    INDEX
    Explanations

    ; punctuation; so

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

    explanation of qualifiers

    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_31_width_262k_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
    可能会
    0.61
    可能會
    0.47
     feels
    0.44
     feel
    0.42
     maybe
    0.39
     possibly
    0.39
     improve
    0.38
    7
    0.38
    （
    0.38
     sometimes
    0.38
    POSITIVE LOGITS
    दरअसल
    0.54
     चूंकि
    0.54
     prvo
    0.51
     दिक्कत
    0.49
     दरअसल
    0.49
     çünkü
    0.46
     هیڅ
    0.46
     বানানো
    0.46
     изначально
    0.46
     dépour
    0.45
    Activations Density 0.016%

    No Known Activations