© 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-65K
    7. 30013
    Prev
    Next
    INDEX
    Explanations

    first, initial

    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_65k_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
    ͖
    0.42
     vlad
    0.41
     mitigate
    0.40
     मिठ
    0.40
    軽減
    0.40
    ädt
    0.39
     skillset
    0.39
     أص
    0.38
    pecia
    0.38
     باع
    0.38
    POSITIVE LOGITS
     First
    0.57
    First
    0.53
    开
    0.53
     Initial
    0.50
    Initial
    0.49
     initial
    0.47
     нача
    0.47
    opening
    0.47
    initial
    0.46
     opening
    0.46
    Activations Density 0.248%

    No Known Activations