© 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. 128444
    Prev
    Next
    INDEX
    Explanations

    purpose and prioritization

    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
     கடுமையான
    0.47
     Eigenschaften
    0.45
     Για
    0.45
    캣
    0.45
     香
    0.44
     हेतु
    0.44
    <unused60>
    0.43
     false
    0.43
     FALSE
    0.43
     parametros
    0.43
    POSITIVE LOGITS
    Recovery
    0.46
    ാൽ
    0.41
     fatta
    0.41
    Americans
    0.40
    栄
    0.40
    ACE
    0.40
     angepasst
    0.40
    Giant
    0.40
    Joint
    0.40
    िवी
    0.40
    Activations Density 0.003%

    No Known Activations