© 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. 128334
    Prev
    Next
    INDEX
    Explanations

    create a self-reinforcing cycle

    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
     Etiquetas
    0.52
    燙
    0.49
    ழ்த்த
    0.48
     Corrections
    0.47
     Hewlett
    0.46
    abilités
    0.46
    paragraphs
    0.45
    fonts
    0.45
    correction
    0.45
     Foundry
    0.44
    POSITIVE LOGITS
    ist
    0.54
    HAN
    0.44
    ANG
    0.44
     outperform
    0.44
     생
    0.43
    ni
    0.43
     astrocyte
    0.43
    MA
    0.43
     symphony
    0.43
    AN
    0.42
    Activations Density 0.004%

    No Known Activations