© 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-IT
    5. 31-GEMMASCOPE-2-RES-262K
    7. 91116
    Prev
    Next
    INDEX
    Explanations

    skill matters and subjective customization

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    un
    0.46
     finished
    0.46
    u
    0.45
     have
    0.43
     reminded
    0.41
     paternal
    0.41
    ين
    0.40
     trustworthy
    0.40
     apresent
    0.40
    io
    0.40
    POSITIVE LOGITS
    污染
    0.50
    LAT
    0.50
     तिलावत
    0.49
    enderung
    0.47
    味道
    0.46
    涛
    0.45
    CH
    0.44
     ค
    0.44
     Fleurit
    0.44
    DEFGHIJKLMNOP
    0.44
    Activations Density 0.012%

    No Known Activations