© 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. 40-GEMMASCOPE-2-RES-65K
    7. 48834
    Prev
    Next
    INDEX
    Explanations

    difficult to do

    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_40_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
     implementation
    0.64
     Implementation
    0.57
     operation
    0.49
     implementations
    0.49
    易
    0.49
    implementation
    0.47
    入手
    0.47
     retrieval
    0.46
     Implement
    0.46
     implement
    0.45
    POSITIVE LOGITS
     kwaliteit
    0.45
    onsen
    0.44
    ielleicht
    0.43
     ອ
    0.42
    埒
    0.42
    avorable
    0.41
    မဲ့
    0.41
    칫
    0.40
    qualität
    0.40
    ulf
    0.40
    Activations Density 0.034%

    No Known Activations